🍻 cheers
Browse files- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Jun26_22-43-33_bernard/events.out.tfevents.1656281899.bernard.1152349.2 +3 -0
- train_results.json +7 -0
- trainer_state.json +3550 -0
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 8.0,
|
3 |
+
"eval_accuracy": 0.912875867386276,
|
4 |
+
"eval_loss": 0.39511793851852417,
|
5 |
+
"eval_runtime": 42.5687,
|
6 |
+
"eval_samples_per_second": 60.937,
|
7 |
+
"eval_steps_per_second": 7.635,
|
8 |
+
"train_loss": 0.4943873079753071,
|
9 |
+
"train_runtime": 5596.588,
|
10 |
+
"train_samples_per_second": 14.595,
|
11 |
+
"train_steps_per_second": 0.913
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 8.0,
|
3 |
+
"eval_accuracy": 0.912875867386276,
|
4 |
+
"eval_loss": 0.39511793851852417,
|
5 |
+
"eval_runtime": 42.5687,
|
6 |
+
"eval_samples_per_second": 60.937,
|
7 |
+
"eval_steps_per_second": 7.635
|
8 |
+
}
|
runs/Jun26_22-43-33_bernard/events.out.tfevents.1656281899.bernard.1152349.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45964b4de3e3c6f5c806af44a61c65f285003de7408bf857e984db5b5903c4db
|
3 |
+
size 363
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 8.0,
|
3 |
+
"train_loss": 0.4943873079753071,
|
4 |
+
"train_runtime": 5596.588,
|
5 |
+
"train_samples_per_second": 14.595,
|
6 |
+
"train_steps_per_second": 0.913
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3550 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.39511793851852417,
|
3 |
+
"best_model_checkpoint": "./exper_batch_16_e8/checkpoint-4900",
|
4 |
+
"epoch": 8.0,
|
5 |
+
"global_step": 5112,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.02,
|
12 |
+
"learning_rate": 0.00019960876369327075,
|
13 |
+
"loss": 4.9227,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.03,
|
18 |
+
"learning_rate": 0.0001992175273865415,
|
19 |
+
"loss": 4.809,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.05,
|
24 |
+
"learning_rate": 0.0001988262910798122,
|
25 |
+
"loss": 4.6411,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.06,
|
30 |
+
"learning_rate": 0.00019843505477308294,
|
31 |
+
"loss": 4.4483,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.08,
|
36 |
+
"learning_rate": 0.00019804381846635368,
|
37 |
+
"loss": 4.3374,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.09,
|
42 |
+
"learning_rate": 0.00019765258215962445,
|
43 |
+
"loss": 4.2538,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.11,
|
48 |
+
"learning_rate": 0.00019726134585289516,
|
49 |
+
"loss": 4.1378,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.13,
|
54 |
+
"learning_rate": 0.0001968701095461659,
|
55 |
+
"loss": 4.0433,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.14,
|
60 |
+
"learning_rate": 0.00019647887323943664,
|
61 |
+
"loss": 4.0425,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.16,
|
66 |
+
"learning_rate": 0.00019608763693270738,
|
67 |
+
"loss": 3.8115,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.16,
|
72 |
+
"eval_accuracy": 0.18619892058596763,
|
73 |
+
"eval_loss": 3.7948496341705322,
|
74 |
+
"eval_runtime": 43.0404,
|
75 |
+
"eval_samples_per_second": 60.269,
|
76 |
+
"eval_steps_per_second": 7.551,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.17,
|
81 |
+
"learning_rate": 0.0001956964006259781,
|
82 |
+
"loss": 3.8373,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.19,
|
87 |
+
"learning_rate": 0.00019530516431924883,
|
88 |
+
"loss": 3.6742,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.2,
|
93 |
+
"learning_rate": 0.00019491392801251957,
|
94 |
+
"loss": 3.581,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.22,
|
99 |
+
"learning_rate": 0.0001945226917057903,
|
100 |
+
"loss": 3.4363,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.23,
|
105 |
+
"learning_rate": 0.00019413145539906105,
|
106 |
+
"loss": 3.433,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.25,
|
111 |
+
"learning_rate": 0.00019374021909233179,
|
112 |
+
"loss": 3.3784,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.27,
|
117 |
+
"learning_rate": 0.00019334898278560253,
|
118 |
+
"loss": 3.2506,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.28,
|
123 |
+
"learning_rate": 0.00019295774647887326,
|
124 |
+
"loss": 3.1149,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.3,
|
129 |
+
"learning_rate": 0.00019256651017214398,
|
130 |
+
"loss": 3.0718,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.31,
|
135 |
+
"learning_rate": 0.00019217527386541472,
|
136 |
+
"loss": 3.1194,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.31,
|
141 |
+
"eval_accuracy": 0.32806476484194297,
|
142 |
+
"eval_loss": 3.0119543075561523,
|
143 |
+
"eval_runtime": 42.8934,
|
144 |
+
"eval_samples_per_second": 60.476,
|
145 |
+
"eval_steps_per_second": 7.577,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.33,
|
150 |
+
"learning_rate": 0.00019178403755868546,
|
151 |
+
"loss": 2.8329,
|
152 |
+
"step": 210
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.34,
|
156 |
+
"learning_rate": 0.0001913928012519562,
|
157 |
+
"loss": 2.9553,
|
158 |
+
"step": 220
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.36,
|
162 |
+
"learning_rate": 0.0001910015649452269,
|
163 |
+
"loss": 2.9676,
|
164 |
+
"step": 230
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.38,
|
168 |
+
"learning_rate": 0.00019061032863849765,
|
169 |
+
"loss": 2.7578,
|
170 |
+
"step": 240
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.39,
|
174 |
+
"learning_rate": 0.00019021909233176841,
|
175 |
+
"loss": 2.6164,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.41,
|
180 |
+
"learning_rate": 0.00018982785602503915,
|
181 |
+
"loss": 2.7734,
|
182 |
+
"step": 260
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.42,
|
186 |
+
"learning_rate": 0.00018943661971830987,
|
187 |
+
"loss": 2.601,
|
188 |
+
"step": 270
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.44,
|
192 |
+
"learning_rate": 0.0001890453834115806,
|
193 |
+
"loss": 2.8081,
|
194 |
+
"step": 280
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.45,
|
198 |
+
"learning_rate": 0.00018865414710485134,
|
199 |
+
"loss": 2.5308,
|
200 |
+
"step": 290
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.47,
|
204 |
+
"learning_rate": 0.00018826291079812208,
|
205 |
+
"loss": 2.3703,
|
206 |
+
"step": 300
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.47,
|
210 |
+
"eval_accuracy": 0.4425597532767926,
|
211 |
+
"eval_loss": 2.479144811630249,
|
212 |
+
"eval_runtime": 42.6319,
|
213 |
+
"eval_samples_per_second": 60.846,
|
214 |
+
"eval_steps_per_second": 7.623,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.49,
|
219 |
+
"learning_rate": 0.0001878716744913928,
|
220 |
+
"loss": 2.5485,
|
221 |
+
"step": 310
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.5,
|
225 |
+
"learning_rate": 0.00018748043818466354,
|
226 |
+
"loss": 2.4543,
|
227 |
+
"step": 320
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.52,
|
231 |
+
"learning_rate": 0.00018708920187793427,
|
232 |
+
"loss": 2.3969,
|
233 |
+
"step": 330
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.53,
|
237 |
+
"learning_rate": 0.00018669796557120501,
|
238 |
+
"loss": 2.4463,
|
239 |
+
"step": 340
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.55,
|
243 |
+
"learning_rate": 0.00018630672926447575,
|
244 |
+
"loss": 2.2321,
|
245 |
+
"step": 350
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.56,
|
249 |
+
"learning_rate": 0.0001859154929577465,
|
250 |
+
"loss": 2.1456,
|
251 |
+
"step": 360
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.58,
|
255 |
+
"learning_rate": 0.00018552425665101723,
|
256 |
+
"loss": 2.152,
|
257 |
+
"step": 370
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.59,
|
261 |
+
"learning_rate": 0.00018513302034428797,
|
262 |
+
"loss": 2.1674,
|
263 |
+
"step": 380
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.61,
|
267 |
+
"learning_rate": 0.00018474178403755868,
|
268 |
+
"loss": 2.1503,
|
269 |
+
"step": 390
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.63,
|
273 |
+
"learning_rate": 0.00018435054773082942,
|
274 |
+
"loss": 2.07,
|
275 |
+
"step": 400
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.63,
|
279 |
+
"eval_accuracy": 0.5,
|
280 |
+
"eval_loss": 2.171962261199951,
|
281 |
+
"eval_runtime": 43.3053,
|
282 |
+
"eval_samples_per_second": 59.9,
|
283 |
+
"eval_steps_per_second": 7.505,
|
284 |
+
"step": 400
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.64,
|
288 |
+
"learning_rate": 0.00018395931142410016,
|
289 |
+
"loss": 2.1113,
|
290 |
+
"step": 410
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.66,
|
294 |
+
"learning_rate": 0.0001835680751173709,
|
295 |
+
"loss": 2.0248,
|
296 |
+
"step": 420
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.67,
|
300 |
+
"learning_rate": 0.00018317683881064161,
|
301 |
+
"loss": 1.9278,
|
302 |
+
"step": 430
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.69,
|
306 |
+
"learning_rate": 0.00018278560250391235,
|
307 |
+
"loss": 1.8649,
|
308 |
+
"step": 440
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.7,
|
312 |
+
"learning_rate": 0.00018239436619718312,
|
313 |
+
"loss": 1.9462,
|
314 |
+
"step": 450
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.72,
|
318 |
+
"learning_rate": 0.00018200312989045386,
|
319 |
+
"loss": 2.0817,
|
320 |
+
"step": 460
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.74,
|
324 |
+
"learning_rate": 0.00018161189358372457,
|
325 |
+
"loss": 1.8968,
|
326 |
+
"step": 470
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.75,
|
330 |
+
"learning_rate": 0.0001812206572769953,
|
331 |
+
"loss": 1.8535,
|
332 |
+
"step": 480
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.77,
|
336 |
+
"learning_rate": 0.00018082942097026605,
|
337 |
+
"loss": 1.8609,
|
338 |
+
"step": 490
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.78,
|
342 |
+
"learning_rate": 0.0001804381846635368,
|
343 |
+
"loss": 1.6847,
|
344 |
+
"step": 500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.78,
|
348 |
+
"eval_accuracy": 0.5956052428681573,
|
349 |
+
"eval_loss": 1.7290887832641602,
|
350 |
+
"eval_runtime": 43.1676,
|
351 |
+
"eval_samples_per_second": 60.091,
|
352 |
+
"eval_steps_per_second": 7.529,
|
353 |
+
"step": 500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.8,
|
357 |
+
"learning_rate": 0.0001800469483568075,
|
358 |
+
"loss": 1.6108,
|
359 |
+
"step": 510
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.81,
|
363 |
+
"learning_rate": 0.00017965571205007824,
|
364 |
+
"loss": 1.6568,
|
365 |
+
"step": 520
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 0.83,
|
369 |
+
"learning_rate": 0.00017926447574334898,
|
370 |
+
"loss": 1.6968,
|
371 |
+
"step": 530
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 0.85,
|
375 |
+
"learning_rate": 0.00017887323943661972,
|
376 |
+
"loss": 1.5102,
|
377 |
+
"step": 540
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.86,
|
381 |
+
"learning_rate": 0.00017848200312989046,
|
382 |
+
"loss": 1.5518,
|
383 |
+
"step": 550
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 0.88,
|
387 |
+
"learning_rate": 0.0001780907668231612,
|
388 |
+
"loss": 1.6318,
|
389 |
+
"step": 560
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.89,
|
393 |
+
"learning_rate": 0.00017769953051643194,
|
394 |
+
"loss": 1.4528,
|
395 |
+
"step": 570
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.91,
|
399 |
+
"learning_rate": 0.00017730829420970268,
|
400 |
+
"loss": 1.5259,
|
401 |
+
"step": 580
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.92,
|
405 |
+
"learning_rate": 0.00017691705790297342,
|
406 |
+
"loss": 1.3553,
|
407 |
+
"step": 590
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 0.94,
|
411 |
+
"learning_rate": 0.00017652582159624413,
|
412 |
+
"loss": 1.3821,
|
413 |
+
"step": 600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.94,
|
417 |
+
"eval_accuracy": 0.6299151888974557,
|
418 |
+
"eval_loss": 1.4777374267578125,
|
419 |
+
"eval_runtime": 42.6804,
|
420 |
+
"eval_samples_per_second": 60.777,
|
421 |
+
"eval_steps_per_second": 7.615,
|
422 |
+
"step": 600
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.95,
|
426 |
+
"learning_rate": 0.00017613458528951487,
|
427 |
+
"loss": 1.5701,
|
428 |
+
"step": 610
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.97,
|
432 |
+
"learning_rate": 0.0001757433489827856,
|
433 |
+
"loss": 1.4866,
|
434 |
+
"step": 620
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.99,
|
438 |
+
"learning_rate": 0.00017535211267605635,
|
439 |
+
"loss": 1.4713,
|
440 |
+
"step": 630
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 1.0,
|
444 |
+
"learning_rate": 0.0001749608763693271,
|
445 |
+
"loss": 1.2937,
|
446 |
+
"step": 640
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 1.02,
|
450 |
+
"learning_rate": 0.00017456964006259783,
|
451 |
+
"loss": 1.169,
|
452 |
+
"step": 650
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 1.03,
|
456 |
+
"learning_rate": 0.00017417840375586857,
|
457 |
+
"loss": 1.1436,
|
458 |
+
"step": 660
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.05,
|
462 |
+
"learning_rate": 0.0001737871674491393,
|
463 |
+
"loss": 1.1583,
|
464 |
+
"step": 670
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 1.06,
|
468 |
+
"learning_rate": 0.00017339593114241002,
|
469 |
+
"loss": 0.9302,
|
470 |
+
"step": 680
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 1.08,
|
474 |
+
"learning_rate": 0.00017300469483568076,
|
475 |
+
"loss": 1.113,
|
476 |
+
"step": 690
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 1.1,
|
480 |
+
"learning_rate": 0.0001726134585289515,
|
481 |
+
"loss": 0.9498,
|
482 |
+
"step": 700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 1.1,
|
486 |
+
"eval_accuracy": 0.6680801850424055,
|
487 |
+
"eval_loss": 1.2934883832931519,
|
488 |
+
"eval_runtime": 42.6632,
|
489 |
+
"eval_samples_per_second": 60.802,
|
490 |
+
"eval_steps_per_second": 7.618,
|
491 |
+
"step": 700
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 1.11,
|
495 |
+
"learning_rate": 0.00017222222222222224,
|
496 |
+
"loss": 0.9322,
|
497 |
+
"step": 710
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 1.13,
|
501 |
+
"learning_rate": 0.00017183098591549295,
|
502 |
+
"loss": 0.9719,
|
503 |
+
"step": 720
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 1.14,
|
507 |
+
"learning_rate": 0.0001714397496087637,
|
508 |
+
"loss": 0.9679,
|
509 |
+
"step": 730
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 1.16,
|
513 |
+
"learning_rate": 0.00017104851330203443,
|
514 |
+
"loss": 0.8539,
|
515 |
+
"step": 740
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 1.17,
|
519 |
+
"learning_rate": 0.0001706572769953052,
|
520 |
+
"loss": 0.971,
|
521 |
+
"step": 750
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.19,
|
525 |
+
"learning_rate": 0.0001702660406885759,
|
526 |
+
"loss": 0.9654,
|
527 |
+
"step": 760
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 1.21,
|
531 |
+
"learning_rate": 0.00016987480438184665,
|
532 |
+
"loss": 0.7825,
|
533 |
+
"step": 770
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 1.22,
|
537 |
+
"learning_rate": 0.00016948356807511739,
|
538 |
+
"loss": 1.0415,
|
539 |
+
"step": 780
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 1.24,
|
543 |
+
"learning_rate": 0.00016909233176838813,
|
544 |
+
"loss": 0.9053,
|
545 |
+
"step": 790
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 1.25,
|
549 |
+
"learning_rate": 0.00016870109546165884,
|
550 |
+
"loss": 0.8741,
|
551 |
+
"step": 800
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 1.25,
|
555 |
+
"eval_accuracy": 0.70508866615266,
|
556 |
+
"eval_loss": 1.135292649269104,
|
557 |
+
"eval_runtime": 42.8607,
|
558 |
+
"eval_samples_per_second": 60.522,
|
559 |
+
"eval_steps_per_second": 7.583,
|
560 |
+
"step": 800
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 1.27,
|
564 |
+
"learning_rate": 0.00016830985915492958,
|
565 |
+
"loss": 0.98,
|
566 |
+
"step": 810
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 1.28,
|
570 |
+
"learning_rate": 0.00016791862284820032,
|
571 |
+
"loss": 0.964,
|
572 |
+
"step": 820
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 1.3,
|
576 |
+
"learning_rate": 0.00016752738654147106,
|
577 |
+
"loss": 1.0007,
|
578 |
+
"step": 830
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 1.31,
|
582 |
+
"learning_rate": 0.0001671361502347418,
|
583 |
+
"loss": 0.7954,
|
584 |
+
"step": 840
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.33,
|
588 |
+
"learning_rate": 0.00016674491392801253,
|
589 |
+
"loss": 0.9806,
|
590 |
+
"step": 850
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 1.35,
|
594 |
+
"learning_rate": 0.00016635367762128327,
|
595 |
+
"loss": 0.8854,
|
596 |
+
"step": 860
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 1.36,
|
600 |
+
"learning_rate": 0.00016596244131455401,
|
601 |
+
"loss": 0.9212,
|
602 |
+
"step": 870
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 1.38,
|
606 |
+
"learning_rate": 0.00016557120500782473,
|
607 |
+
"loss": 0.8027,
|
608 |
+
"step": 880
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 1.39,
|
612 |
+
"learning_rate": 0.00016517996870109547,
|
613 |
+
"loss": 0.7901,
|
614 |
+
"step": 890
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 1.41,
|
618 |
+
"learning_rate": 0.0001647887323943662,
|
619 |
+
"loss": 0.8875,
|
620 |
+
"step": 900
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 1.41,
|
624 |
+
"eval_accuracy": 0.7447956823438705,
|
625 |
+
"eval_loss": 0.9951499700546265,
|
626 |
+
"eval_runtime": 42.5213,
|
627 |
+
"eval_samples_per_second": 61.005,
|
628 |
+
"eval_steps_per_second": 7.643,
|
629 |
+
"step": 900
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 1.42,
|
633 |
+
"learning_rate": 0.00016439749608763694,
|
634 |
+
"loss": 0.8624,
|
635 |
+
"step": 910
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 1.44,
|
639 |
+
"learning_rate": 0.00016400625978090766,
|
640 |
+
"loss": 0.7194,
|
641 |
+
"step": 920
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 1.46,
|
645 |
+
"learning_rate": 0.0001636150234741784,
|
646 |
+
"loss": 0.8012,
|
647 |
+
"step": 930
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.47,
|
651 |
+
"learning_rate": 0.00016322378716744916,
|
652 |
+
"loss": 0.9272,
|
653 |
+
"step": 940
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 1.49,
|
657 |
+
"learning_rate": 0.0001628325508607199,
|
658 |
+
"loss": 0.8623,
|
659 |
+
"step": 950
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 1.5,
|
663 |
+
"learning_rate": 0.00016244131455399061,
|
664 |
+
"loss": 0.8576,
|
665 |
+
"step": 960
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 1.52,
|
669 |
+
"learning_rate": 0.00016205007824726135,
|
670 |
+
"loss": 0.6199,
|
671 |
+
"step": 970
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 1.53,
|
675 |
+
"learning_rate": 0.0001616588419405321,
|
676 |
+
"loss": 0.6498,
|
677 |
+
"step": 980
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 1.55,
|
681 |
+
"learning_rate": 0.00016126760563380283,
|
682 |
+
"loss": 0.8226,
|
683 |
+
"step": 990
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 1.56,
|
687 |
+
"learning_rate": 0.00016087636932707354,
|
688 |
+
"loss": 0.7233,
|
689 |
+
"step": 1000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.56,
|
693 |
+
"eval_accuracy": 0.748650732459522,
|
694 |
+
"eval_loss": 0.926507294178009,
|
695 |
+
"eval_runtime": 43.2867,
|
696 |
+
"eval_samples_per_second": 59.926,
|
697 |
+
"eval_steps_per_second": 7.508,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 1.58,
|
702 |
+
"learning_rate": 0.00016048513302034428,
|
703 |
+
"loss": 0.7018,
|
704 |
+
"step": 1010
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 1.6,
|
708 |
+
"learning_rate": 0.00016009389671361502,
|
709 |
+
"loss": 0.739,
|
710 |
+
"step": 1020
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.61,
|
714 |
+
"learning_rate": 0.00015970266040688576,
|
715 |
+
"loss": 0.6795,
|
716 |
+
"step": 1030
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 1.63,
|
720 |
+
"learning_rate": 0.0001593114241001565,
|
721 |
+
"loss": 0.6543,
|
722 |
+
"step": 1040
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 1.64,
|
726 |
+
"learning_rate": 0.00015892018779342724,
|
727 |
+
"loss": 0.7367,
|
728 |
+
"step": 1050
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 1.66,
|
732 |
+
"learning_rate": 0.00015852895148669798,
|
733 |
+
"loss": 0.6326,
|
734 |
+
"step": 1060
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 1.67,
|
738 |
+
"learning_rate": 0.00015813771517996872,
|
739 |
+
"loss": 0.5459,
|
740 |
+
"step": 1070
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 1.69,
|
744 |
+
"learning_rate": 0.00015774647887323943,
|
745 |
+
"loss": 0.8587,
|
746 |
+
"step": 1080
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 1.71,
|
750 |
+
"learning_rate": 0.00015735524256651017,
|
751 |
+
"loss": 0.6657,
|
752 |
+
"step": 1090
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.72,
|
756 |
+
"learning_rate": 0.0001569640062597809,
|
757 |
+
"loss": 0.6696,
|
758 |
+
"step": 1100
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 1.72,
|
762 |
+
"eval_accuracy": 0.7625289128758674,
|
763 |
+
"eval_loss": 0.8659528493881226,
|
764 |
+
"eval_runtime": 42.5957,
|
765 |
+
"eval_samples_per_second": 60.898,
|
766 |
+
"eval_steps_per_second": 7.63,
|
767 |
+
"step": 1100
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 1.74,
|
771 |
+
"learning_rate": 0.00015657276995305165,
|
772 |
+
"loss": 0.7146,
|
773 |
+
"step": 1110
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.75,
|
777 |
+
"learning_rate": 0.00015618153364632236,
|
778 |
+
"loss": 0.4619,
|
779 |
+
"step": 1120
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 1.77,
|
783 |
+
"learning_rate": 0.0001557902973395931,
|
784 |
+
"loss": 0.6367,
|
785 |
+
"step": 1130
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 1.78,
|
789 |
+
"learning_rate": 0.00015539906103286387,
|
790 |
+
"loss": 0.585,
|
791 |
+
"step": 1140
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 1.8,
|
795 |
+
"learning_rate": 0.0001550078247261346,
|
796 |
+
"loss": 0.5464,
|
797 |
+
"step": 1150
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 1.82,
|
801 |
+
"learning_rate": 0.00015461658841940532,
|
802 |
+
"loss": 0.7695,
|
803 |
+
"step": 1160
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 1.83,
|
807 |
+
"learning_rate": 0.00015422535211267606,
|
808 |
+
"loss": 0.7019,
|
809 |
+
"step": 1170
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 1.85,
|
813 |
+
"learning_rate": 0.0001538341158059468,
|
814 |
+
"loss": 0.4508,
|
815 |
+
"step": 1180
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.86,
|
819 |
+
"learning_rate": 0.00015344287949921754,
|
820 |
+
"loss": 0.6396,
|
821 |
+
"step": 1190
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 1.88,
|
825 |
+
"learning_rate": 0.00015305164319248828,
|
826 |
+
"loss": 0.7364,
|
827 |
+
"step": 1200
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 1.88,
|
831 |
+
"eval_accuracy": 0.7579028527370856,
|
832 |
+
"eval_loss": 0.8710386753082275,
|
833 |
+
"eval_runtime": 42.6201,
|
834 |
+
"eval_samples_per_second": 60.863,
|
835 |
+
"eval_steps_per_second": 7.626,
|
836 |
+
"step": 1200
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.89,
|
840 |
+
"learning_rate": 0.000152660406885759,
|
841 |
+
"loss": 0.8301,
|
842 |
+
"step": 1210
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 1.91,
|
846 |
+
"learning_rate": 0.00015226917057902973,
|
847 |
+
"loss": 0.6424,
|
848 |
+
"step": 1220
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 1.92,
|
852 |
+
"learning_rate": 0.00015187793427230047,
|
853 |
+
"loss": 0.788,
|
854 |
+
"step": 1230
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.94,
|
858 |
+
"learning_rate": 0.0001514866979655712,
|
859 |
+
"loss": 0.8115,
|
860 |
+
"step": 1240
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 1.96,
|
864 |
+
"learning_rate": 0.00015109546165884195,
|
865 |
+
"loss": 0.7002,
|
866 |
+
"step": 1250
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 1.97,
|
870 |
+
"learning_rate": 0.0001507042253521127,
|
871 |
+
"loss": 0.5162,
|
872 |
+
"step": 1260
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 1.99,
|
876 |
+
"learning_rate": 0.00015031298904538343,
|
877 |
+
"loss": 0.5706,
|
878 |
+
"step": 1270
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 2.0,
|
882 |
+
"learning_rate": 0.00014992175273865417,
|
883 |
+
"loss": 0.4565,
|
884 |
+
"step": 1280
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 2.02,
|
888 |
+
"learning_rate": 0.00014953051643192488,
|
889 |
+
"loss": 0.3758,
|
890 |
+
"step": 1290
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 2.03,
|
894 |
+
"learning_rate": 0.00014913928012519562,
|
895 |
+
"loss": 0.3933,
|
896 |
+
"step": 1300
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 2.03,
|
900 |
+
"eval_accuracy": 0.8037779491133384,
|
901 |
+
"eval_loss": 0.7161900997161865,
|
902 |
+
"eval_runtime": 42.7187,
|
903 |
+
"eval_samples_per_second": 60.723,
|
904 |
+
"eval_steps_per_second": 7.608,
|
905 |
+
"step": 1300
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 2.05,
|
909 |
+
"learning_rate": 0.00014874804381846636,
|
910 |
+
"loss": 0.2625,
|
911 |
+
"step": 1310
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 2.07,
|
915 |
+
"learning_rate": 0.0001483568075117371,
|
916 |
+
"loss": 0.3585,
|
917 |
+
"step": 1320
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 2.08,
|
921 |
+
"learning_rate": 0.00014796557120500784,
|
922 |
+
"loss": 0.3735,
|
923 |
+
"step": 1330
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 2.1,
|
927 |
+
"learning_rate": 0.00014757433489827858,
|
928 |
+
"loss": 0.4221,
|
929 |
+
"step": 1340
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 2.11,
|
933 |
+
"learning_rate": 0.00014718309859154932,
|
934 |
+
"loss": 0.3249,
|
935 |
+
"step": 1350
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 2.13,
|
939 |
+
"learning_rate": 0.00014679186228482006,
|
940 |
+
"loss": 0.4802,
|
941 |
+
"step": 1360
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 2.14,
|
945 |
+
"learning_rate": 0.00014640062597809077,
|
946 |
+
"loss": 0.3311,
|
947 |
+
"step": 1370
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 2.16,
|
951 |
+
"learning_rate": 0.0001460093896713615,
|
952 |
+
"loss": 0.4276,
|
953 |
+
"step": 1380
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 2.18,
|
957 |
+
"learning_rate": 0.00014561815336463225,
|
958 |
+
"loss": 0.222,
|
959 |
+
"step": 1390
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 2.19,
|
963 |
+
"learning_rate": 0.00014522691705790299,
|
964 |
+
"loss": 0.3443,
|
965 |
+
"step": 1400
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 2.19,
|
969 |
+
"eval_accuracy": 0.8299922898997687,
|
970 |
+
"eval_loss": 0.6305217742919922,
|
971 |
+
"eval_runtime": 43.2454,
|
972 |
+
"eval_samples_per_second": 59.983,
|
973 |
+
"eval_steps_per_second": 7.515,
|
974 |
+
"step": 1400
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 2.21,
|
978 |
+
"learning_rate": 0.0001448356807511737,
|
979 |
+
"loss": 0.3609,
|
980 |
+
"step": 1410
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 2.22,
|
984 |
+
"learning_rate": 0.00014444444444444444,
|
985 |
+
"loss": 0.4267,
|
986 |
+
"step": 1420
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 2.24,
|
990 |
+
"learning_rate": 0.00014405320813771518,
|
991 |
+
"loss": 0.4206,
|
992 |
+
"step": 1430
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 2.25,
|
996 |
+
"learning_rate": 0.00014366197183098594,
|
997 |
+
"loss": 0.3414,
|
998 |
+
"step": 1440
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 2.27,
|
1002 |
+
"learning_rate": 0.00014327073552425666,
|
1003 |
+
"loss": 0.3351,
|
1004 |
+
"step": 1450
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 2.28,
|
1008 |
+
"learning_rate": 0.0001428794992175274,
|
1009 |
+
"loss": 0.2785,
|
1010 |
+
"step": 1460
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 2.3,
|
1014 |
+
"learning_rate": 0.00014248826291079813,
|
1015 |
+
"loss": 0.2454,
|
1016 |
+
"step": 1470
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 2.32,
|
1020 |
+
"learning_rate": 0.00014209702660406887,
|
1021 |
+
"loss": 0.4382,
|
1022 |
+
"step": 1480
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 2.33,
|
1026 |
+
"learning_rate": 0.0001417057902973396,
|
1027 |
+
"loss": 0.4117,
|
1028 |
+
"step": 1490
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 2.35,
|
1032 |
+
"learning_rate": 0.00014131455399061033,
|
1033 |
+
"loss": 0.3376,
|
1034 |
+
"step": 1500
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 2.35,
|
1038 |
+
"eval_accuracy": 0.8315343099460293,
|
1039 |
+
"eval_loss": 0.62732994556427,
|
1040 |
+
"eval_runtime": 43.1453,
|
1041 |
+
"eval_samples_per_second": 60.122,
|
1042 |
+
"eval_steps_per_second": 7.533,
|
1043 |
+
"step": 1500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 2.36,
|
1047 |
+
"learning_rate": 0.00014092331768388107,
|
1048 |
+
"loss": 0.4258,
|
1049 |
+
"step": 1510
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 2.38,
|
1053 |
+
"learning_rate": 0.0001405320813771518,
|
1054 |
+
"loss": 0.3927,
|
1055 |
+
"step": 1520
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 2.39,
|
1059 |
+
"learning_rate": 0.00014014084507042254,
|
1060 |
+
"loss": 0.4155,
|
1061 |
+
"step": 1530
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 2.41,
|
1065 |
+
"learning_rate": 0.00013974960876369328,
|
1066 |
+
"loss": 0.2391,
|
1067 |
+
"step": 1540
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 2.43,
|
1071 |
+
"learning_rate": 0.00013935837245696402,
|
1072 |
+
"loss": 0.2629,
|
1073 |
+
"step": 1550
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 2.44,
|
1077 |
+
"learning_rate": 0.00013896713615023476,
|
1078 |
+
"loss": 0.3786,
|
1079 |
+
"step": 1560
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 2.46,
|
1083 |
+
"learning_rate": 0.00013857589984350547,
|
1084 |
+
"loss": 0.3195,
|
1085 |
+
"step": 1570
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 2.47,
|
1089 |
+
"learning_rate": 0.00013818466353677621,
|
1090 |
+
"loss": 0.2218,
|
1091 |
+
"step": 1580
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 2.49,
|
1095 |
+
"learning_rate": 0.00013779342723004695,
|
1096 |
+
"loss": 0.2454,
|
1097 |
+
"step": 1590
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 2.5,
|
1101 |
+
"learning_rate": 0.0001374021909233177,
|
1102 |
+
"loss": 0.3071,
|
1103 |
+
"step": 1600
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 2.5,
|
1107 |
+
"eval_accuracy": 0.8319198149575945,
|
1108 |
+
"eval_loss": 0.5988024473190308,
|
1109 |
+
"eval_runtime": 43.1472,
|
1110 |
+
"eval_samples_per_second": 60.12,
|
1111 |
+
"eval_steps_per_second": 7.532,
|
1112 |
+
"step": 1600
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 2.52,
|
1116 |
+
"learning_rate": 0.0001370109546165884,
|
1117 |
+
"loss": 0.2436,
|
1118 |
+
"step": 1610
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 2.54,
|
1122 |
+
"learning_rate": 0.00013661971830985914,
|
1123 |
+
"loss": 0.244,
|
1124 |
+
"step": 1620
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 2.55,
|
1128 |
+
"learning_rate": 0.0001362284820031299,
|
1129 |
+
"loss": 0.3613,
|
1130 |
+
"step": 1630
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 2.57,
|
1134 |
+
"learning_rate": 0.00013583724569640065,
|
1135 |
+
"loss": 0.4596,
|
1136 |
+
"step": 1640
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 2.58,
|
1140 |
+
"learning_rate": 0.00013544600938967136,
|
1141 |
+
"loss": 0.2966,
|
1142 |
+
"step": 1650
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 2.6,
|
1146 |
+
"learning_rate": 0.0001350547730829421,
|
1147 |
+
"loss": 0.3579,
|
1148 |
+
"step": 1660
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 2.61,
|
1152 |
+
"learning_rate": 0.00013466353677621284,
|
1153 |
+
"loss": 0.3748,
|
1154 |
+
"step": 1670
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 2.63,
|
1158 |
+
"learning_rate": 0.00013427230046948358,
|
1159 |
+
"loss": 0.2644,
|
1160 |
+
"step": 1680
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 2.64,
|
1164 |
+
"learning_rate": 0.0001338810641627543,
|
1165 |
+
"loss": 0.2401,
|
1166 |
+
"step": 1690
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 2.66,
|
1170 |
+
"learning_rate": 0.00013348982785602503,
|
1171 |
+
"loss": 0.2863,
|
1172 |
+
"step": 1700
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 2.66,
|
1176 |
+
"eval_accuracy": 0.8153430994602929,
|
1177 |
+
"eval_loss": 0.6730513572692871,
|
1178 |
+
"eval_runtime": 42.369,
|
1179 |
+
"eval_samples_per_second": 61.224,
|
1180 |
+
"eval_steps_per_second": 7.671,
|
1181 |
+
"step": 1700
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 2.68,
|
1185 |
+
"learning_rate": 0.00013309859154929577,
|
1186 |
+
"loss": 0.1875,
|
1187 |
+
"step": 1710
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 2.69,
|
1191 |
+
"learning_rate": 0.0001327073552425665,
|
1192 |
+
"loss": 0.2504,
|
1193 |
+
"step": 1720
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 2.71,
|
1197 |
+
"learning_rate": 0.00013231611893583725,
|
1198 |
+
"loss": 0.3071,
|
1199 |
+
"step": 1730
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 2.72,
|
1203 |
+
"learning_rate": 0.000131924882629108,
|
1204 |
+
"loss": 0.3549,
|
1205 |
+
"step": 1740
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 2.74,
|
1209 |
+
"learning_rate": 0.00013153364632237873,
|
1210 |
+
"loss": 0.3535,
|
1211 |
+
"step": 1750
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 2.75,
|
1215 |
+
"learning_rate": 0.00013114241001564947,
|
1216 |
+
"loss": 0.3758,
|
1217 |
+
"step": 1760
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 2.77,
|
1221 |
+
"learning_rate": 0.0001307511737089202,
|
1222 |
+
"loss": 0.3174,
|
1223 |
+
"step": 1770
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 2.79,
|
1227 |
+
"learning_rate": 0.00013035993740219092,
|
1228 |
+
"loss": 0.2612,
|
1229 |
+
"step": 1780
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.8,
|
1233 |
+
"learning_rate": 0.00012996870109546166,
|
1234 |
+
"loss": 0.3562,
|
1235 |
+
"step": 1790
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 2.82,
|
1239 |
+
"learning_rate": 0.0001295774647887324,
|
1240 |
+
"loss": 0.3017,
|
1241 |
+
"step": 1800
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 2.82,
|
1245 |
+
"eval_accuracy": 0.8315343099460293,
|
1246 |
+
"eval_loss": 0.6042011976242065,
|
1247 |
+
"eval_runtime": 43.2556,
|
1248 |
+
"eval_samples_per_second": 59.969,
|
1249 |
+
"eval_steps_per_second": 7.513,
|
1250 |
+
"step": 1800
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.83,
|
1254 |
+
"learning_rate": 0.00012918622848200314,
|
1255 |
+
"loss": 0.4068,
|
1256 |
+
"step": 1810
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 2.85,
|
1260 |
+
"learning_rate": 0.00012879499217527385,
|
1261 |
+
"loss": 0.2903,
|
1262 |
+
"step": 1820
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 2.86,
|
1266 |
+
"learning_rate": 0.00012840375586854462,
|
1267 |
+
"loss": 0.2385,
|
1268 |
+
"step": 1830
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 2.88,
|
1272 |
+
"learning_rate": 0.00012801251956181536,
|
1273 |
+
"loss": 0.2733,
|
1274 |
+
"step": 1840
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 2.9,
|
1278 |
+
"learning_rate": 0.0001276212832550861,
|
1279 |
+
"loss": 0.2174,
|
1280 |
+
"step": 1850
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 2.91,
|
1284 |
+
"learning_rate": 0.0001272300469483568,
|
1285 |
+
"loss": 0.2813,
|
1286 |
+
"step": 1860
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 2.93,
|
1290 |
+
"learning_rate": 0.00012683881064162755,
|
1291 |
+
"loss": 0.1874,
|
1292 |
+
"step": 1870
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.94,
|
1296 |
+
"learning_rate": 0.0001264475743348983,
|
1297 |
+
"loss": 0.3027,
|
1298 |
+
"step": 1880
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 2.96,
|
1302 |
+
"learning_rate": 0.00012605633802816903,
|
1303 |
+
"loss": 0.3872,
|
1304 |
+
"step": 1890
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 2.97,
|
1308 |
+
"learning_rate": 0.00012566510172143974,
|
1309 |
+
"loss": 0.2382,
|
1310 |
+
"step": 1900
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 2.97,
|
1314 |
+
"eval_accuracy": 0.8712413261372398,
|
1315 |
+
"eval_loss": 0.5117685794830322,
|
1316 |
+
"eval_runtime": 43.2378,
|
1317 |
+
"eval_samples_per_second": 59.994,
|
1318 |
+
"eval_steps_per_second": 7.517,
|
1319 |
+
"step": 1900
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 2.99,
|
1323 |
+
"learning_rate": 0.00012527386541471048,
|
1324 |
+
"loss": 0.2799,
|
1325 |
+
"step": 1910
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 3.0,
|
1329 |
+
"learning_rate": 0.00012488262910798122,
|
1330 |
+
"loss": 0.2348,
|
1331 |
+
"step": 1920
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 3.02,
|
1335 |
+
"learning_rate": 0.00012449139280125196,
|
1336 |
+
"loss": 0.1834,
|
1337 |
+
"step": 1930
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 3.04,
|
1341 |
+
"learning_rate": 0.0001241001564945227,
|
1342 |
+
"loss": 0.2012,
|
1343 |
+
"step": 1940
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 3.05,
|
1347 |
+
"learning_rate": 0.00012370892018779344,
|
1348 |
+
"loss": 0.182,
|
1349 |
+
"step": 1950
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 3.07,
|
1353 |
+
"learning_rate": 0.00012331768388106418,
|
1354 |
+
"loss": 0.1467,
|
1355 |
+
"step": 1960
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 3.08,
|
1359 |
+
"learning_rate": 0.00012292644757433492,
|
1360 |
+
"loss": 0.2158,
|
1361 |
+
"step": 1970
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 3.1,
|
1365 |
+
"learning_rate": 0.00012253521126760563,
|
1366 |
+
"loss": 0.1666,
|
1367 |
+
"step": 1980
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 3.11,
|
1371 |
+
"learning_rate": 0.00012214397496087637,
|
1372 |
+
"loss": 0.1223,
|
1373 |
+
"step": 1990
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 3.13,
|
1377 |
+
"learning_rate": 0.00012175273865414711,
|
1378 |
+
"loss": 0.1578,
|
1379 |
+
"step": 2000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 3.13,
|
1383 |
+
"eval_accuracy": 0.8735543562066307,
|
1384 |
+
"eval_loss": 0.49165645241737366,
|
1385 |
+
"eval_runtime": 42.6596,
|
1386 |
+
"eval_samples_per_second": 60.807,
|
1387 |
+
"eval_steps_per_second": 7.618,
|
1388 |
+
"step": 2000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 3.15,
|
1392 |
+
"learning_rate": 0.00012136150234741786,
|
1393 |
+
"loss": 0.2114,
|
1394 |
+
"step": 2010
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 3.16,
|
1398 |
+
"learning_rate": 0.00012097026604068857,
|
1399 |
+
"loss": 0.1657,
|
1400 |
+
"step": 2020
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 3.18,
|
1404 |
+
"learning_rate": 0.00012057902973395931,
|
1405 |
+
"loss": 0.1228,
|
1406 |
+
"step": 2030
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 3.19,
|
1410 |
+
"learning_rate": 0.00012018779342723005,
|
1411 |
+
"loss": 0.1246,
|
1412 |
+
"step": 2040
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 3.21,
|
1416 |
+
"learning_rate": 0.00011979655712050079,
|
1417 |
+
"loss": 0.1231,
|
1418 |
+
"step": 2050
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 3.22,
|
1422 |
+
"learning_rate": 0.00011940532081377152,
|
1423 |
+
"loss": 0.2011,
|
1424 |
+
"step": 2060
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 3.24,
|
1428 |
+
"learning_rate": 0.00011901408450704226,
|
1429 |
+
"loss": 0.1177,
|
1430 |
+
"step": 2070
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 3.26,
|
1434 |
+
"learning_rate": 0.000118622848200313,
|
1435 |
+
"loss": 0.1258,
|
1436 |
+
"step": 2080
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 3.27,
|
1440 |
+
"learning_rate": 0.00011823161189358373,
|
1441 |
+
"loss": 0.1668,
|
1442 |
+
"step": 2090
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 3.29,
|
1446 |
+
"learning_rate": 0.00011784037558685446,
|
1447 |
+
"loss": 0.1794,
|
1448 |
+
"step": 2100
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 3.29,
|
1452 |
+
"eval_accuracy": 0.8631457208943716,
|
1453 |
+
"eval_loss": 0.53019118309021,
|
1454 |
+
"eval_runtime": 43.3542,
|
1455 |
+
"eval_samples_per_second": 59.833,
|
1456 |
+
"eval_steps_per_second": 7.496,
|
1457 |
+
"step": 2100
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 3.3,
|
1461 |
+
"learning_rate": 0.0001174491392801252,
|
1462 |
+
"loss": 0.1714,
|
1463 |
+
"step": 2110
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 3.32,
|
1467 |
+
"learning_rate": 0.00011705790297339594,
|
1468 |
+
"loss": 0.0935,
|
1469 |
+
"step": 2120
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 3.33,
|
1473 |
+
"learning_rate": 0.00011666666666666668,
|
1474 |
+
"loss": 0.1422,
|
1475 |
+
"step": 2130
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 3.35,
|
1479 |
+
"learning_rate": 0.0001162754303599374,
|
1480 |
+
"loss": 0.1965,
|
1481 |
+
"step": 2140
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 3.36,
|
1485 |
+
"learning_rate": 0.00011588419405320814,
|
1486 |
+
"loss": 0.1014,
|
1487 |
+
"step": 2150
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 3.38,
|
1491 |
+
"learning_rate": 0.00011549295774647888,
|
1492 |
+
"loss": 0.2306,
|
1493 |
+
"step": 2160
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 3.4,
|
1497 |
+
"learning_rate": 0.00011510172143974962,
|
1498 |
+
"loss": 0.1482,
|
1499 |
+
"step": 2170
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 3.41,
|
1503 |
+
"learning_rate": 0.00011471048513302034,
|
1504 |
+
"loss": 0.1835,
|
1505 |
+
"step": 2180
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 3.43,
|
1509 |
+
"learning_rate": 0.00011431924882629107,
|
1510 |
+
"loss": 0.0886,
|
1511 |
+
"step": 2190
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 3.44,
|
1515 |
+
"learning_rate": 0.00011392801251956183,
|
1516 |
+
"loss": 0.1093,
|
1517 |
+
"step": 2200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 3.44,
|
1521 |
+
"eval_accuracy": 0.8635312259059368,
|
1522 |
+
"eval_loss": 0.5034898519515991,
|
1523 |
+
"eval_runtime": 42.4689,
|
1524 |
+
"eval_samples_per_second": 61.08,
|
1525 |
+
"eval_steps_per_second": 7.653,
|
1526 |
+
"step": 2200
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 3.46,
|
1530 |
+
"learning_rate": 0.00011353677621283257,
|
1531 |
+
"loss": 0.12,
|
1532 |
+
"step": 2210
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 3.47,
|
1536 |
+
"learning_rate": 0.00011314553990610328,
|
1537 |
+
"loss": 0.1595,
|
1538 |
+
"step": 2220
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 3.49,
|
1542 |
+
"learning_rate": 0.00011275430359937402,
|
1543 |
+
"loss": 0.1279,
|
1544 |
+
"step": 2230
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 3.51,
|
1548 |
+
"learning_rate": 0.00011236306729264476,
|
1549 |
+
"loss": 0.1693,
|
1550 |
+
"step": 2240
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 3.52,
|
1554 |
+
"learning_rate": 0.0001119718309859155,
|
1555 |
+
"loss": 0.1546,
|
1556 |
+
"step": 2250
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 3.54,
|
1560 |
+
"learning_rate": 0.00011158059467918622,
|
1561 |
+
"loss": 0.0954,
|
1562 |
+
"step": 2260
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 3.55,
|
1566 |
+
"learning_rate": 0.00011118935837245696,
|
1567 |
+
"loss": 0.1307,
|
1568 |
+
"step": 2270
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 3.57,
|
1572 |
+
"learning_rate": 0.0001107981220657277,
|
1573 |
+
"loss": 0.1041,
|
1574 |
+
"step": 2280
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 3.58,
|
1578 |
+
"learning_rate": 0.00011040688575899844,
|
1579 |
+
"loss": 0.0915,
|
1580 |
+
"step": 2290
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 3.6,
|
1584 |
+
"learning_rate": 0.00011001564945226917,
|
1585 |
+
"loss": 0.1076,
|
1586 |
+
"step": 2300
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 3.6,
|
1590 |
+
"eval_accuracy": 0.8673862760215882,
|
1591 |
+
"eval_loss": 0.5186420679092407,
|
1592 |
+
"eval_runtime": 42.5346,
|
1593 |
+
"eval_samples_per_second": 60.986,
|
1594 |
+
"eval_steps_per_second": 7.641,
|
1595 |
+
"step": 2300
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 3.62,
|
1599 |
+
"learning_rate": 0.00010962441314553991,
|
1600 |
+
"loss": 0.1075,
|
1601 |
+
"step": 2310
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 3.63,
|
1605 |
+
"learning_rate": 0.00010923317683881065,
|
1606 |
+
"loss": 0.0914,
|
1607 |
+
"step": 2320
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 3.65,
|
1611 |
+
"learning_rate": 0.00010884194053208139,
|
1612 |
+
"loss": 0.1057,
|
1613 |
+
"step": 2330
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 3.66,
|
1617 |
+
"learning_rate": 0.00010845070422535213,
|
1618 |
+
"loss": 0.1967,
|
1619 |
+
"step": 2340
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 3.68,
|
1623 |
+
"learning_rate": 0.00010805946791862285,
|
1624 |
+
"loss": 0.1148,
|
1625 |
+
"step": 2350
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 3.69,
|
1629 |
+
"learning_rate": 0.00010766823161189359,
|
1630 |
+
"loss": 0.0591,
|
1631 |
+
"step": 2360
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 3.71,
|
1635 |
+
"learning_rate": 0.00010727699530516433,
|
1636 |
+
"loss": 0.1369,
|
1637 |
+
"step": 2370
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 3.72,
|
1641 |
+
"learning_rate": 0.00010688575899843507,
|
1642 |
+
"loss": 0.0991,
|
1643 |
+
"step": 2380
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 3.74,
|
1647 |
+
"learning_rate": 0.0001064945226917058,
|
1648 |
+
"loss": 0.1579,
|
1649 |
+
"step": 2390
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 3.76,
|
1653 |
+
"learning_rate": 0.00010610328638497653,
|
1654 |
+
"loss": 0.1219,
|
1655 |
+
"step": 2400
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 3.76,
|
1659 |
+
"eval_accuracy": 0.8801079414032382,
|
1660 |
+
"eval_loss": 0.4722863733768463,
|
1661 |
+
"eval_runtime": 42.9174,
|
1662 |
+
"eval_samples_per_second": 60.442,
|
1663 |
+
"eval_steps_per_second": 7.573,
|
1664 |
+
"step": 2400
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 3.77,
|
1668 |
+
"learning_rate": 0.00010571205007824727,
|
1669 |
+
"loss": 0.0743,
|
1670 |
+
"step": 2410
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 3.79,
|
1674 |
+
"learning_rate": 0.00010532081377151801,
|
1675 |
+
"loss": 0.1099,
|
1676 |
+
"step": 2420
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 3.8,
|
1680 |
+
"learning_rate": 0.00010492957746478873,
|
1681 |
+
"loss": 0.1275,
|
1682 |
+
"step": 2430
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 3.82,
|
1686 |
+
"learning_rate": 0.00010453834115805947,
|
1687 |
+
"loss": 0.0914,
|
1688 |
+
"step": 2440
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 3.83,
|
1692 |
+
"learning_rate": 0.00010414710485133022,
|
1693 |
+
"loss": 0.1177,
|
1694 |
+
"step": 2450
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 3.85,
|
1698 |
+
"learning_rate": 0.00010375586854460096,
|
1699 |
+
"loss": 0.1255,
|
1700 |
+
"step": 2460
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 3.87,
|
1704 |
+
"learning_rate": 0.00010336463223787167,
|
1705 |
+
"loss": 0.1376,
|
1706 |
+
"step": 2470
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 3.88,
|
1710 |
+
"learning_rate": 0.00010297339593114241,
|
1711 |
+
"loss": 0.1848,
|
1712 |
+
"step": 2480
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 3.9,
|
1716 |
+
"learning_rate": 0.00010258215962441315,
|
1717 |
+
"loss": 0.0986,
|
1718 |
+
"step": 2490
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 3.91,
|
1722 |
+
"learning_rate": 0.0001021909233176839,
|
1723 |
+
"loss": 0.1017,
|
1724 |
+
"step": 2500
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 3.91,
|
1728 |
+
"eval_accuracy": 0.8712413261372398,
|
1729 |
+
"eval_loss": 0.5132078528404236,
|
1730 |
+
"eval_runtime": 42.9886,
|
1731 |
+
"eval_samples_per_second": 60.342,
|
1732 |
+
"eval_steps_per_second": 7.56,
|
1733 |
+
"step": 2500
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 3.93,
|
1737 |
+
"learning_rate": 0.00010179968701095461,
|
1738 |
+
"loss": 0.1514,
|
1739 |
+
"step": 2510
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 3.94,
|
1743 |
+
"learning_rate": 0.00010140845070422535,
|
1744 |
+
"loss": 0.0884,
|
1745 |
+
"step": 2520
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 3.96,
|
1749 |
+
"learning_rate": 0.00010101721439749609,
|
1750 |
+
"loss": 0.0894,
|
1751 |
+
"step": 2530
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 3.97,
|
1755 |
+
"learning_rate": 0.00010062597809076683,
|
1756 |
+
"loss": 0.0964,
|
1757 |
+
"step": 2540
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 3.99,
|
1761 |
+
"learning_rate": 0.00010023474178403756,
|
1762 |
+
"loss": 0.0851,
|
1763 |
+
"step": 2550
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 4.01,
|
1767 |
+
"learning_rate": 9.98435054773083e-05,
|
1768 |
+
"loss": 0.078,
|
1769 |
+
"step": 2560
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 4.02,
|
1773 |
+
"learning_rate": 9.945226917057904e-05,
|
1774 |
+
"loss": 0.0611,
|
1775 |
+
"step": 2570
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 4.04,
|
1779 |
+
"learning_rate": 9.906103286384976e-05,
|
1780 |
+
"loss": 0.0721,
|
1781 |
+
"step": 2580
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 4.05,
|
1785 |
+
"learning_rate": 9.86697965571205e-05,
|
1786 |
+
"loss": 0.0604,
|
1787 |
+
"step": 2590
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 4.07,
|
1791 |
+
"learning_rate": 9.827856025039124e-05,
|
1792 |
+
"loss": 0.0351,
|
1793 |
+
"step": 2600
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 4.07,
|
1797 |
+
"eval_accuracy": 0.8727833461835004,
|
1798 |
+
"eval_loss": 0.4709137976169586,
|
1799 |
+
"eval_runtime": 43.1107,
|
1800 |
+
"eval_samples_per_second": 60.171,
|
1801 |
+
"eval_steps_per_second": 7.539,
|
1802 |
+
"step": 2600
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 4.08,
|
1806 |
+
"learning_rate": 9.788732394366198e-05,
|
1807 |
+
"loss": 0.0262,
|
1808 |
+
"step": 2610
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 4.1,
|
1812 |
+
"learning_rate": 9.749608763693271e-05,
|
1813 |
+
"loss": 0.0263,
|
1814 |
+
"step": 2620
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 4.12,
|
1818 |
+
"learning_rate": 9.710485133020345e-05,
|
1819 |
+
"loss": 0.0201,
|
1820 |
+
"step": 2630
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 4.13,
|
1824 |
+
"learning_rate": 9.671361502347419e-05,
|
1825 |
+
"loss": 0.0467,
|
1826 |
+
"step": 2640
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 4.15,
|
1830 |
+
"learning_rate": 9.632237871674493e-05,
|
1831 |
+
"loss": 0.0635,
|
1832 |
+
"step": 2650
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 4.16,
|
1836 |
+
"learning_rate": 9.593114241001565e-05,
|
1837 |
+
"loss": 0.0162,
|
1838 |
+
"step": 2660
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 4.18,
|
1842 |
+
"learning_rate": 9.553990610328639e-05,
|
1843 |
+
"loss": 0.1138,
|
1844 |
+
"step": 2670
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 4.19,
|
1848 |
+
"learning_rate": 9.514866979655712e-05,
|
1849 |
+
"loss": 0.0597,
|
1850 |
+
"step": 2680
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 4.21,
|
1854 |
+
"learning_rate": 9.475743348982786e-05,
|
1855 |
+
"loss": 0.0349,
|
1856 |
+
"step": 2690
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 4.23,
|
1860 |
+
"learning_rate": 9.43661971830986e-05,
|
1861 |
+
"loss": 0.0295,
|
1862 |
+
"step": 2700
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 4.23,
|
1866 |
+
"eval_accuracy": 0.8824209714726291,
|
1867 |
+
"eval_loss": 0.4673934876918793,
|
1868 |
+
"eval_runtime": 43.0785,
|
1869 |
+
"eval_samples_per_second": 60.216,
|
1870 |
+
"eval_steps_per_second": 7.544,
|
1871 |
+
"step": 2700
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 4.24,
|
1875 |
+
"learning_rate": 9.397496087636933e-05,
|
1876 |
+
"loss": 0.0375,
|
1877 |
+
"step": 2710
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 4.26,
|
1881 |
+
"learning_rate": 9.358372456964006e-05,
|
1882 |
+
"loss": 0.0238,
|
1883 |
+
"step": 2720
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 4.27,
|
1887 |
+
"learning_rate": 9.31924882629108e-05,
|
1888 |
+
"loss": 0.0262,
|
1889 |
+
"step": 2730
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 4.29,
|
1893 |
+
"learning_rate": 9.280125195618154e-05,
|
1894 |
+
"loss": 0.0507,
|
1895 |
+
"step": 2740
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 4.3,
|
1899 |
+
"learning_rate": 9.241001564945228e-05,
|
1900 |
+
"loss": 0.0397,
|
1901 |
+
"step": 2750
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 4.32,
|
1905 |
+
"learning_rate": 9.2018779342723e-05,
|
1906 |
+
"loss": 0.0486,
|
1907 |
+
"step": 2760
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 4.33,
|
1911 |
+
"learning_rate": 9.162754303599374e-05,
|
1912 |
+
"loss": 0.0332,
|
1913 |
+
"step": 2770
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 4.35,
|
1917 |
+
"learning_rate": 9.123630672926447e-05,
|
1918 |
+
"loss": 0.022,
|
1919 |
+
"step": 2780
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 4.37,
|
1923 |
+
"learning_rate": 9.084507042253522e-05,
|
1924 |
+
"loss": 0.0735,
|
1925 |
+
"step": 2790
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 4.38,
|
1929 |
+
"learning_rate": 9.045383411580595e-05,
|
1930 |
+
"loss": 0.0416,
|
1931 |
+
"step": 2800
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 4.38,
|
1935 |
+
"eval_accuracy": 0.8804934464148034,
|
1936 |
+
"eval_loss": 0.4836331307888031,
|
1937 |
+
"eval_runtime": 42.924,
|
1938 |
+
"eval_samples_per_second": 60.432,
|
1939 |
+
"eval_steps_per_second": 7.572,
|
1940 |
+
"step": 2800
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 4.4,
|
1944 |
+
"learning_rate": 9.006259780907669e-05,
|
1945 |
+
"loss": 0.0188,
|
1946 |
+
"step": 2810
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 4.41,
|
1950 |
+
"learning_rate": 8.967136150234741e-05,
|
1951 |
+
"loss": 0.0361,
|
1952 |
+
"step": 2820
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 4.43,
|
1956 |
+
"learning_rate": 8.928012519561815e-05,
|
1957 |
+
"loss": 0.0655,
|
1958 |
+
"step": 2830
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 4.44,
|
1962 |
+
"learning_rate": 8.888888888888889e-05,
|
1963 |
+
"loss": 0.097,
|
1964 |
+
"step": 2840
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 4.46,
|
1968 |
+
"learning_rate": 8.849765258215963e-05,
|
1969 |
+
"loss": 0.1409,
|
1970 |
+
"step": 2850
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 4.48,
|
1974 |
+
"learning_rate": 8.810641627543036e-05,
|
1975 |
+
"loss": 0.0353,
|
1976 |
+
"step": 2860
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 4.49,
|
1980 |
+
"learning_rate": 8.77151799687011e-05,
|
1981 |
+
"loss": 0.036,
|
1982 |
+
"step": 2870
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 4.51,
|
1986 |
+
"learning_rate": 8.732394366197182e-05,
|
1987 |
+
"loss": 0.0322,
|
1988 |
+
"step": 2880
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 4.52,
|
1992 |
+
"learning_rate": 8.693270735524258e-05,
|
1993 |
+
"loss": 0.1002,
|
1994 |
+
"step": 2890
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 4.54,
|
1998 |
+
"learning_rate": 8.65414710485133e-05,
|
1999 |
+
"loss": 0.0386,
|
2000 |
+
"step": 2900
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 4.54,
|
2004 |
+
"eval_accuracy": 0.8828064764841943,
|
2005 |
+
"eval_loss": 0.466279000043869,
|
2006 |
+
"eval_runtime": 43.3083,
|
2007 |
+
"eval_samples_per_second": 59.896,
|
2008 |
+
"eval_steps_per_second": 7.504,
|
2009 |
+
"step": 2900
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 4.55,
|
2013 |
+
"learning_rate": 8.615023474178404e-05,
|
2014 |
+
"loss": 0.0219,
|
2015 |
+
"step": 2910
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 4.57,
|
2019 |
+
"learning_rate": 8.575899843505478e-05,
|
2020 |
+
"loss": 0.0428,
|
2021 |
+
"step": 2920
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 4.59,
|
2025 |
+
"learning_rate": 8.536776212832551e-05,
|
2026 |
+
"loss": 0.0941,
|
2027 |
+
"step": 2930
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 4.6,
|
2031 |
+
"learning_rate": 8.497652582159625e-05,
|
2032 |
+
"loss": 0.061,
|
2033 |
+
"step": 2940
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 4.62,
|
2037 |
+
"learning_rate": 8.458528951486699e-05,
|
2038 |
+
"loss": 0.0524,
|
2039 |
+
"step": 2950
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 4.63,
|
2043 |
+
"learning_rate": 8.419405320813773e-05,
|
2044 |
+
"loss": 0.0961,
|
2045 |
+
"step": 2960
|
2046 |
+
},
|
2047 |
+
{
|
2048 |
+
"epoch": 4.65,
|
2049 |
+
"learning_rate": 8.380281690140845e-05,
|
2050 |
+
"loss": 0.0524,
|
2051 |
+
"step": 2970
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 4.66,
|
2055 |
+
"learning_rate": 8.341158059467919e-05,
|
2056 |
+
"loss": 0.0604,
|
2057 |
+
"step": 2980
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 4.68,
|
2061 |
+
"learning_rate": 8.302034428794993e-05,
|
2062 |
+
"loss": 0.0534,
|
2063 |
+
"step": 2990
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 4.69,
|
2067 |
+
"learning_rate": 8.262910798122067e-05,
|
2068 |
+
"loss": 0.0392,
|
2069 |
+
"step": 3000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 4.69,
|
2073 |
+
"eval_accuracy": 0.8989976869699307,
|
2074 |
+
"eval_loss": 0.4003235101699829,
|
2075 |
+
"eval_runtime": 43.2036,
|
2076 |
+
"eval_samples_per_second": 60.041,
|
2077 |
+
"eval_steps_per_second": 7.523,
|
2078 |
+
"step": 3000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 4.71,
|
2082 |
+
"learning_rate": 8.22378716744914e-05,
|
2083 |
+
"loss": 0.0273,
|
2084 |
+
"step": 3010
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 4.73,
|
2088 |
+
"learning_rate": 8.184663536776213e-05,
|
2089 |
+
"loss": 0.0529,
|
2090 |
+
"step": 3020
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 4.74,
|
2094 |
+
"learning_rate": 8.145539906103286e-05,
|
2095 |
+
"loss": 0.0726,
|
2096 |
+
"step": 3030
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 4.76,
|
2100 |
+
"learning_rate": 8.106416275430361e-05,
|
2101 |
+
"loss": 0.0505,
|
2102 |
+
"step": 3040
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 4.77,
|
2106 |
+
"learning_rate": 8.067292644757434e-05,
|
2107 |
+
"loss": 0.0778,
|
2108 |
+
"step": 3050
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 4.79,
|
2112 |
+
"learning_rate": 8.028169014084508e-05,
|
2113 |
+
"loss": 0.0891,
|
2114 |
+
"step": 3060
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 4.8,
|
2118 |
+
"learning_rate": 7.98904538341158e-05,
|
2119 |
+
"loss": 0.0264,
|
2120 |
+
"step": 3070
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 4.82,
|
2124 |
+
"learning_rate": 7.949921752738654e-05,
|
2125 |
+
"loss": 0.0112,
|
2126 |
+
"step": 3080
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 4.84,
|
2130 |
+
"learning_rate": 7.910798122065728e-05,
|
2131 |
+
"loss": 0.0163,
|
2132 |
+
"step": 3090
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 4.85,
|
2136 |
+
"learning_rate": 7.871674491392802e-05,
|
2137 |
+
"loss": 0.0383,
|
2138 |
+
"step": 3100
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 4.85,
|
2142 |
+
"eval_accuracy": 0.894757131842714,
|
2143 |
+
"eval_loss": 0.4187348186969757,
|
2144 |
+
"eval_runtime": 43.069,
|
2145 |
+
"eval_samples_per_second": 60.229,
|
2146 |
+
"eval_steps_per_second": 7.546,
|
2147 |
+
"step": 3100
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 4.87,
|
2151 |
+
"learning_rate": 7.832550860719875e-05,
|
2152 |
+
"loss": 0.0867,
|
2153 |
+
"step": 3110
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 4.88,
|
2157 |
+
"learning_rate": 7.793427230046949e-05,
|
2158 |
+
"loss": 0.0649,
|
2159 |
+
"step": 3120
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 4.9,
|
2163 |
+
"learning_rate": 7.754303599374021e-05,
|
2164 |
+
"loss": 0.0959,
|
2165 |
+
"step": 3130
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 4.91,
|
2169 |
+
"learning_rate": 7.715179968701097e-05,
|
2170 |
+
"loss": 0.0634,
|
2171 |
+
"step": 3140
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 4.93,
|
2175 |
+
"learning_rate": 7.676056338028169e-05,
|
2176 |
+
"loss": 0.0607,
|
2177 |
+
"step": 3150
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 4.95,
|
2181 |
+
"learning_rate": 7.636932707355243e-05,
|
2182 |
+
"loss": 0.0246,
|
2183 |
+
"step": 3160
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 4.96,
|
2187 |
+
"learning_rate": 7.597809076682316e-05,
|
2188 |
+
"loss": 0.0421,
|
2189 |
+
"step": 3170
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 4.98,
|
2193 |
+
"learning_rate": 7.55868544600939e-05,
|
2194 |
+
"loss": 0.0324,
|
2195 |
+
"step": 3180
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 4.99,
|
2199 |
+
"learning_rate": 7.519561815336464e-05,
|
2200 |
+
"loss": 0.0119,
|
2201 |
+
"step": 3190
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 5.01,
|
2205 |
+
"learning_rate": 7.480438184663538e-05,
|
2206 |
+
"loss": 0.0624,
|
2207 |
+
"step": 3200
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 5.01,
|
2211 |
+
"eval_accuracy": 0.887432536622976,
|
2212 |
+
"eval_loss": 0.44595664739608765,
|
2213 |
+
"eval_runtime": 42.8626,
|
2214 |
+
"eval_samples_per_second": 60.519,
|
2215 |
+
"eval_steps_per_second": 7.582,
|
2216 |
+
"step": 3200
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 5.02,
|
2220 |
+
"learning_rate": 7.44131455399061e-05,
|
2221 |
+
"loss": 0.0158,
|
2222 |
+
"step": 3210
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 5.04,
|
2226 |
+
"learning_rate": 7.402190923317684e-05,
|
2227 |
+
"loss": 0.0264,
|
2228 |
+
"step": 3220
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 5.05,
|
2232 |
+
"learning_rate": 7.363067292644758e-05,
|
2233 |
+
"loss": 0.0129,
|
2234 |
+
"step": 3230
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 5.07,
|
2238 |
+
"learning_rate": 7.323943661971832e-05,
|
2239 |
+
"loss": 0.0153,
|
2240 |
+
"step": 3240
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 5.09,
|
2244 |
+
"learning_rate": 7.284820031298905e-05,
|
2245 |
+
"loss": 0.0111,
|
2246 |
+
"step": 3250
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 5.1,
|
2250 |
+
"learning_rate": 7.245696400625979e-05,
|
2251 |
+
"loss": 0.0179,
|
2252 |
+
"step": 3260
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 5.12,
|
2256 |
+
"learning_rate": 7.206572769953051e-05,
|
2257 |
+
"loss": 0.0174,
|
2258 |
+
"step": 3270
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 5.13,
|
2262 |
+
"learning_rate": 7.167449139280125e-05,
|
2263 |
+
"loss": 0.0396,
|
2264 |
+
"step": 3280
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 5.15,
|
2268 |
+
"learning_rate": 7.128325508607199e-05,
|
2269 |
+
"loss": 0.0073,
|
2270 |
+
"step": 3290
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 5.16,
|
2274 |
+
"learning_rate": 7.089201877934273e-05,
|
2275 |
+
"loss": 0.0188,
|
2276 |
+
"step": 3300
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 5.16,
|
2280 |
+
"eval_accuracy": 0.9028527370855821,
|
2281 |
+
"eval_loss": 0.41694527864456177,
|
2282 |
+
"eval_runtime": 42.4867,
|
2283 |
+
"eval_samples_per_second": 61.054,
|
2284 |
+
"eval_steps_per_second": 7.649,
|
2285 |
+
"step": 3300
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 5.18,
|
2289 |
+
"learning_rate": 7.050078247261346e-05,
|
2290 |
+
"loss": 0.0048,
|
2291 |
+
"step": 3310
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 5.2,
|
2295 |
+
"learning_rate": 7.01095461658842e-05,
|
2296 |
+
"loss": 0.006,
|
2297 |
+
"step": 3320
|
2298 |
+
},
|
2299 |
+
{
|
2300 |
+
"epoch": 5.21,
|
2301 |
+
"learning_rate": 6.971830985915493e-05,
|
2302 |
+
"loss": 0.018,
|
2303 |
+
"step": 3330
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 5.23,
|
2307 |
+
"learning_rate": 6.932707355242567e-05,
|
2308 |
+
"loss": 0.0314,
|
2309 |
+
"step": 3340
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 5.24,
|
2313 |
+
"learning_rate": 6.89358372456964e-05,
|
2314 |
+
"loss": 0.0113,
|
2315 |
+
"step": 3350
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 5.26,
|
2319 |
+
"learning_rate": 6.854460093896714e-05,
|
2320 |
+
"loss": 0.0075,
|
2321 |
+
"step": 3360
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 5.27,
|
2325 |
+
"learning_rate": 6.815336463223787e-05,
|
2326 |
+
"loss": 0.0148,
|
2327 |
+
"step": 3370
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 5.29,
|
2331 |
+
"learning_rate": 6.77621283255086e-05,
|
2332 |
+
"loss": 0.0092,
|
2333 |
+
"step": 3380
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 5.31,
|
2337 |
+
"learning_rate": 6.737089201877934e-05,
|
2338 |
+
"loss": 0.022,
|
2339 |
+
"step": 3390
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 5.32,
|
2343 |
+
"learning_rate": 6.697965571205008e-05,
|
2344 |
+
"loss": 0.0174,
|
2345 |
+
"step": 3400
|
2346 |
+
},
|
2347 |
+
{
|
2348 |
+
"epoch": 5.32,
|
2349 |
+
"eval_accuracy": 0.8951426368542791,
|
2350 |
+
"eval_loss": 0.40977469086647034,
|
2351 |
+
"eval_runtime": 42.8425,
|
2352 |
+
"eval_samples_per_second": 60.547,
|
2353 |
+
"eval_steps_per_second": 7.586,
|
2354 |
+
"step": 3400
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 5.34,
|
2358 |
+
"learning_rate": 6.658841940532081e-05,
|
2359 |
+
"loss": 0.0054,
|
2360 |
+
"step": 3410
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 5.35,
|
2364 |
+
"learning_rate": 6.619718309859155e-05,
|
2365 |
+
"loss": 0.0051,
|
2366 |
+
"step": 3420
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 5.37,
|
2370 |
+
"learning_rate": 6.580594679186229e-05,
|
2371 |
+
"loss": 0.0316,
|
2372 |
+
"step": 3430
|
2373 |
+
},
|
2374 |
+
{
|
2375 |
+
"epoch": 5.38,
|
2376 |
+
"learning_rate": 6.541471048513303e-05,
|
2377 |
+
"loss": 0.0127,
|
2378 |
+
"step": 3440
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 5.4,
|
2382 |
+
"learning_rate": 6.502347417840375e-05,
|
2383 |
+
"loss": 0.0085,
|
2384 |
+
"step": 3450
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 5.41,
|
2388 |
+
"learning_rate": 6.463223787167449e-05,
|
2389 |
+
"loss": 0.0383,
|
2390 |
+
"step": 3460
|
2391 |
+
},
|
2392 |
+
{
|
2393 |
+
"epoch": 5.43,
|
2394 |
+
"learning_rate": 6.424100156494522e-05,
|
2395 |
+
"loss": 0.0075,
|
2396 |
+
"step": 3470
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 5.45,
|
2400 |
+
"learning_rate": 6.384976525821597e-05,
|
2401 |
+
"loss": 0.0049,
|
2402 |
+
"step": 3480
|
2403 |
+
},
|
2404 |
+
{
|
2405 |
+
"epoch": 5.46,
|
2406 |
+
"learning_rate": 6.345852895148671e-05,
|
2407 |
+
"loss": 0.0259,
|
2408 |
+
"step": 3490
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 5.48,
|
2412 |
+
"learning_rate": 6.306729264475744e-05,
|
2413 |
+
"loss": 0.0257,
|
2414 |
+
"step": 3500
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 5.48,
|
2418 |
+
"eval_accuracy": 0.8951426368542791,
|
2419 |
+
"eval_loss": 0.4288833737373352,
|
2420 |
+
"eval_runtime": 42.534,
|
2421 |
+
"eval_samples_per_second": 60.987,
|
2422 |
+
"eval_steps_per_second": 7.641,
|
2423 |
+
"step": 3500
|
2424 |
+
},
|
2425 |
+
{
|
2426 |
+
"epoch": 5.49,
|
2427 |
+
"learning_rate": 6.267605633802818e-05,
|
2428 |
+
"loss": 0.0194,
|
2429 |
+
"step": 3510
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 5.51,
|
2433 |
+
"learning_rate": 6.22848200312989e-05,
|
2434 |
+
"loss": 0.0144,
|
2435 |
+
"step": 3520
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 5.52,
|
2439 |
+
"learning_rate": 6.189358372456964e-05,
|
2440 |
+
"loss": 0.0128,
|
2441 |
+
"step": 3530
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 5.54,
|
2445 |
+
"learning_rate": 6.150234741784038e-05,
|
2446 |
+
"loss": 0.0273,
|
2447 |
+
"step": 3540
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 5.56,
|
2451 |
+
"learning_rate": 6.111111111111112e-05,
|
2452 |
+
"loss": 0.0447,
|
2453 |
+
"step": 3550
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 5.57,
|
2457 |
+
"learning_rate": 6.071987480438185e-05,
|
2458 |
+
"loss": 0.0389,
|
2459 |
+
"step": 3560
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 5.59,
|
2463 |
+
"learning_rate": 6.032863849765259e-05,
|
2464 |
+
"loss": 0.0288,
|
2465 |
+
"step": 3570
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 5.6,
|
2469 |
+
"learning_rate": 5.993740219092332e-05,
|
2470 |
+
"loss": 0.0101,
|
2471 |
+
"step": 3580
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 5.62,
|
2475 |
+
"learning_rate": 5.954616588419406e-05,
|
2476 |
+
"loss": 0.006,
|
2477 |
+
"step": 3590
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 5.63,
|
2481 |
+
"learning_rate": 5.915492957746479e-05,
|
2482 |
+
"loss": 0.0123,
|
2483 |
+
"step": 3600
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 5.63,
|
2487 |
+
"eval_accuracy": 0.9028527370855821,
|
2488 |
+
"eval_loss": 0.4294571876525879,
|
2489 |
+
"eval_runtime": 42.3329,
|
2490 |
+
"eval_samples_per_second": 61.276,
|
2491 |
+
"eval_steps_per_second": 7.677,
|
2492 |
+
"step": 3600
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 5.65,
|
2496 |
+
"learning_rate": 5.876369327073553e-05,
|
2497 |
+
"loss": 0.017,
|
2498 |
+
"step": 3610
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 5.67,
|
2502 |
+
"learning_rate": 5.837245696400626e-05,
|
2503 |
+
"loss": 0.0132,
|
2504 |
+
"step": 3620
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 5.68,
|
2508 |
+
"learning_rate": 5.7981220657277e-05,
|
2509 |
+
"loss": 0.0201,
|
2510 |
+
"step": 3630
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 5.7,
|
2514 |
+
"learning_rate": 5.7589984350547735e-05,
|
2515 |
+
"loss": 0.0104,
|
2516 |
+
"step": 3640
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 5.71,
|
2520 |
+
"learning_rate": 5.7198748043818474e-05,
|
2521 |
+
"loss": 0.0054,
|
2522 |
+
"step": 3650
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 5.73,
|
2526 |
+
"learning_rate": 5.68075117370892e-05,
|
2527 |
+
"loss": 0.0137,
|
2528 |
+
"step": 3660
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 5.74,
|
2532 |
+
"learning_rate": 5.6416275430359946e-05,
|
2533 |
+
"loss": 0.0038,
|
2534 |
+
"step": 3670
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 5.76,
|
2538 |
+
"learning_rate": 5.602503912363067e-05,
|
2539 |
+
"loss": 0.0052,
|
2540 |
+
"step": 3680
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 5.77,
|
2544 |
+
"learning_rate": 5.563380281690141e-05,
|
2545 |
+
"loss": 0.0142,
|
2546 |
+
"step": 3690
|
2547 |
+
},
|
2548 |
+
{
|
2549 |
+
"epoch": 5.79,
|
2550 |
+
"learning_rate": 5.5242566510172144e-05,
|
2551 |
+
"loss": 0.0052,
|
2552 |
+
"step": 3700
|
2553 |
+
},
|
2554 |
+
{
|
2555 |
+
"epoch": 5.79,
|
2556 |
+
"eval_accuracy": 0.8993831919814957,
|
2557 |
+
"eval_loss": 0.43951472640037537,
|
2558 |
+
"eval_runtime": 42.973,
|
2559 |
+
"eval_samples_per_second": 60.364,
|
2560 |
+
"eval_steps_per_second": 7.563,
|
2561 |
+
"step": 3700
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 5.81,
|
2565 |
+
"learning_rate": 5.4851330203442884e-05,
|
2566 |
+
"loss": 0.0176,
|
2567 |
+
"step": 3710
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 5.82,
|
2571 |
+
"learning_rate": 5.4460093896713616e-05,
|
2572 |
+
"loss": 0.0226,
|
2573 |
+
"step": 3720
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 5.84,
|
2577 |
+
"learning_rate": 5.4068857589984356e-05,
|
2578 |
+
"loss": 0.007,
|
2579 |
+
"step": 3730
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 5.85,
|
2583 |
+
"learning_rate": 5.367762128325509e-05,
|
2584 |
+
"loss": 0.0444,
|
2585 |
+
"step": 3740
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 5.87,
|
2589 |
+
"learning_rate": 5.328638497652583e-05,
|
2590 |
+
"loss": 0.0056,
|
2591 |
+
"step": 3750
|
2592 |
+
},
|
2593 |
+
{
|
2594 |
+
"epoch": 5.88,
|
2595 |
+
"learning_rate": 5.2895148669796554e-05,
|
2596 |
+
"loss": 0.0172,
|
2597 |
+
"step": 3760
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 5.9,
|
2601 |
+
"learning_rate": 5.25039123630673e-05,
|
2602 |
+
"loss": 0.0235,
|
2603 |
+
"step": 3770
|
2604 |
+
},
|
2605 |
+
{
|
2606 |
+
"epoch": 5.92,
|
2607 |
+
"learning_rate": 5.2112676056338026e-05,
|
2608 |
+
"loss": 0.0065,
|
2609 |
+
"step": 3780
|
2610 |
+
},
|
2611 |
+
{
|
2612 |
+
"epoch": 5.93,
|
2613 |
+
"learning_rate": 5.172143974960877e-05,
|
2614 |
+
"loss": 0.0083,
|
2615 |
+
"step": 3790
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 5.95,
|
2619 |
+
"learning_rate": 5.13302034428795e-05,
|
2620 |
+
"loss": 0.0081,
|
2621 |
+
"step": 3800
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 5.95,
|
2625 |
+
"eval_accuracy": 0.9082498072474943,
|
2626 |
+
"eval_loss": 0.42170995473861694,
|
2627 |
+
"eval_runtime": 43.0345,
|
2628 |
+
"eval_samples_per_second": 60.277,
|
2629 |
+
"eval_steps_per_second": 7.552,
|
2630 |
+
"step": 3800
|
2631 |
+
},
|
2632 |
+
{
|
2633 |
+
"epoch": 5.96,
|
2634 |
+
"learning_rate": 5.093896713615024e-05,
|
2635 |
+
"loss": 0.0071,
|
2636 |
+
"step": 3810
|
2637 |
+
},
|
2638 |
+
{
|
2639 |
+
"epoch": 5.98,
|
2640 |
+
"learning_rate": 5.054773082942097e-05,
|
2641 |
+
"loss": 0.031,
|
2642 |
+
"step": 3820
|
2643 |
+
},
|
2644 |
+
{
|
2645 |
+
"epoch": 5.99,
|
2646 |
+
"learning_rate": 5.015649452269171e-05,
|
2647 |
+
"loss": 0.0057,
|
2648 |
+
"step": 3830
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 6.01,
|
2652 |
+
"learning_rate": 4.976525821596245e-05,
|
2653 |
+
"loss": 0.0051,
|
2654 |
+
"step": 3840
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 6.03,
|
2658 |
+
"learning_rate": 4.937402190923318e-05,
|
2659 |
+
"loss": 0.0064,
|
2660 |
+
"step": 3850
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 6.04,
|
2664 |
+
"learning_rate": 4.8982785602503914e-05,
|
2665 |
+
"loss": 0.0033,
|
2666 |
+
"step": 3860
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 6.06,
|
2670 |
+
"learning_rate": 4.8591549295774653e-05,
|
2671 |
+
"loss": 0.0033,
|
2672 |
+
"step": 3870
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 6.07,
|
2676 |
+
"learning_rate": 4.8200312989045386e-05,
|
2677 |
+
"loss": 0.0032,
|
2678 |
+
"step": 3880
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 6.09,
|
2682 |
+
"learning_rate": 4.7809076682316126e-05,
|
2683 |
+
"loss": 0.0033,
|
2684 |
+
"step": 3890
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 6.1,
|
2688 |
+
"learning_rate": 4.741784037558686e-05,
|
2689 |
+
"loss": 0.0032,
|
2690 |
+
"step": 3900
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 6.1,
|
2694 |
+
"eval_accuracy": 0.9055512721665382,
|
2695 |
+
"eval_loss": 0.42163705825805664,
|
2696 |
+
"eval_runtime": 43.2354,
|
2697 |
+
"eval_samples_per_second": 59.997,
|
2698 |
+
"eval_steps_per_second": 7.517,
|
2699 |
+
"step": 3900
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 6.12,
|
2703 |
+
"learning_rate": 4.702660406885759e-05,
|
2704 |
+
"loss": 0.0035,
|
2705 |
+
"step": 3910
|
2706 |
+
},
|
2707 |
+
{
|
2708 |
+
"epoch": 6.13,
|
2709 |
+
"learning_rate": 4.663536776212833e-05,
|
2710 |
+
"loss": 0.0032,
|
2711 |
+
"step": 3920
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 6.15,
|
2715 |
+
"learning_rate": 4.624413145539906e-05,
|
2716 |
+
"loss": 0.003,
|
2717 |
+
"step": 3930
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 6.17,
|
2721 |
+
"learning_rate": 4.58528951486698e-05,
|
2722 |
+
"loss": 0.0031,
|
2723 |
+
"step": 3940
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 6.18,
|
2727 |
+
"learning_rate": 4.5461658841940535e-05,
|
2728 |
+
"loss": 0.0029,
|
2729 |
+
"step": 3950
|
2730 |
+
},
|
2731 |
+
{
|
2732 |
+
"epoch": 6.2,
|
2733 |
+
"learning_rate": 4.507042253521127e-05,
|
2734 |
+
"loss": 0.0115,
|
2735 |
+
"step": 3960
|
2736 |
+
},
|
2737 |
+
{
|
2738 |
+
"epoch": 6.21,
|
2739 |
+
"learning_rate": 4.467918622848201e-05,
|
2740 |
+
"loss": 0.0037,
|
2741 |
+
"step": 3970
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 6.23,
|
2745 |
+
"learning_rate": 4.428794992175274e-05,
|
2746 |
+
"loss": 0.0056,
|
2747 |
+
"step": 3980
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 6.24,
|
2751 |
+
"learning_rate": 4.389671361502348e-05,
|
2752 |
+
"loss": 0.0027,
|
2753 |
+
"step": 3990
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 6.26,
|
2757 |
+
"learning_rate": 4.350547730829421e-05,
|
2758 |
+
"loss": 0.0033,
|
2759 |
+
"step": 4000
|
2760 |
+
},
|
2761 |
+
{
|
2762 |
+
"epoch": 6.26,
|
2763 |
+
"eval_accuracy": 0.9082498072474943,
|
2764 |
+
"eval_loss": 0.4112599790096283,
|
2765 |
+
"eval_runtime": 42.7966,
|
2766 |
+
"eval_samples_per_second": 60.612,
|
2767 |
+
"eval_steps_per_second": 7.594,
|
2768 |
+
"step": 4000
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 6.28,
|
2772 |
+
"learning_rate": 4.311424100156495e-05,
|
2773 |
+
"loss": 0.0044,
|
2774 |
+
"step": 4010
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 6.29,
|
2778 |
+
"learning_rate": 4.2723004694835684e-05,
|
2779 |
+
"loss": 0.0033,
|
2780 |
+
"step": 4020
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 6.31,
|
2784 |
+
"learning_rate": 4.2331768388106416e-05,
|
2785 |
+
"loss": 0.0031,
|
2786 |
+
"step": 4030
|
2787 |
+
},
|
2788 |
+
{
|
2789 |
+
"epoch": 6.32,
|
2790 |
+
"learning_rate": 4.1940532081377156e-05,
|
2791 |
+
"loss": 0.0033,
|
2792 |
+
"step": 4040
|
2793 |
+
},
|
2794 |
+
{
|
2795 |
+
"epoch": 6.34,
|
2796 |
+
"learning_rate": 4.154929577464789e-05,
|
2797 |
+
"loss": 0.0048,
|
2798 |
+
"step": 4050
|
2799 |
+
},
|
2800 |
+
{
|
2801 |
+
"epoch": 6.35,
|
2802 |
+
"learning_rate": 4.115805946791863e-05,
|
2803 |
+
"loss": 0.003,
|
2804 |
+
"step": 4060
|
2805 |
+
},
|
2806 |
+
{
|
2807 |
+
"epoch": 6.37,
|
2808 |
+
"learning_rate": 4.076682316118936e-05,
|
2809 |
+
"loss": 0.0044,
|
2810 |
+
"step": 4070
|
2811 |
+
},
|
2812 |
+
{
|
2813 |
+
"epoch": 6.38,
|
2814 |
+
"learning_rate": 4.037558685446009e-05,
|
2815 |
+
"loss": 0.0037,
|
2816 |
+
"step": 4080
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 6.4,
|
2820 |
+
"learning_rate": 3.998435054773083e-05,
|
2821 |
+
"loss": 0.0024,
|
2822 |
+
"step": 4090
|
2823 |
+
},
|
2824 |
+
{
|
2825 |
+
"epoch": 6.42,
|
2826 |
+
"learning_rate": 3.9593114241001565e-05,
|
2827 |
+
"loss": 0.0024,
|
2828 |
+
"step": 4100
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 6.42,
|
2832 |
+
"eval_accuracy": 0.9101773323053199,
|
2833 |
+
"eval_loss": 0.40595710277557373,
|
2834 |
+
"eval_runtime": 42.4804,
|
2835 |
+
"eval_samples_per_second": 61.063,
|
2836 |
+
"eval_steps_per_second": 7.651,
|
2837 |
+
"step": 4100
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 6.43,
|
2841 |
+
"learning_rate": 3.9201877934272305e-05,
|
2842 |
+
"loss": 0.0027,
|
2843 |
+
"step": 4110
|
2844 |
+
},
|
2845 |
+
{
|
2846 |
+
"epoch": 6.45,
|
2847 |
+
"learning_rate": 3.881064162754304e-05,
|
2848 |
+
"loss": 0.0036,
|
2849 |
+
"step": 4120
|
2850 |
+
},
|
2851 |
+
{
|
2852 |
+
"epoch": 6.46,
|
2853 |
+
"learning_rate": 3.841940532081377e-05,
|
2854 |
+
"loss": 0.0029,
|
2855 |
+
"step": 4130
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 6.48,
|
2859 |
+
"learning_rate": 3.802816901408451e-05,
|
2860 |
+
"loss": 0.0057,
|
2861 |
+
"step": 4140
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 6.49,
|
2865 |
+
"learning_rate": 3.763693270735524e-05,
|
2866 |
+
"loss": 0.0033,
|
2867 |
+
"step": 4150
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 6.51,
|
2871 |
+
"learning_rate": 3.724569640062598e-05,
|
2872 |
+
"loss": 0.003,
|
2873 |
+
"step": 4160
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 6.53,
|
2877 |
+
"learning_rate": 3.6854460093896714e-05,
|
2878 |
+
"loss": 0.0028,
|
2879 |
+
"step": 4170
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 6.54,
|
2883 |
+
"learning_rate": 3.646322378716745e-05,
|
2884 |
+
"loss": 0.0028,
|
2885 |
+
"step": 4180
|
2886 |
+
},
|
2887 |
+
{
|
2888 |
+
"epoch": 6.56,
|
2889 |
+
"learning_rate": 3.6071987480438186e-05,
|
2890 |
+
"loss": 0.0075,
|
2891 |
+
"step": 4190
|
2892 |
+
},
|
2893 |
+
{
|
2894 |
+
"epoch": 6.57,
|
2895 |
+
"learning_rate": 3.568075117370892e-05,
|
2896 |
+
"loss": 0.0022,
|
2897 |
+
"step": 4200
|
2898 |
+
},
|
2899 |
+
{
|
2900 |
+
"epoch": 6.57,
|
2901 |
+
"eval_accuracy": 0.9090208172706246,
|
2902 |
+
"eval_loss": 0.4067469835281372,
|
2903 |
+
"eval_runtime": 42.6487,
|
2904 |
+
"eval_samples_per_second": 60.823,
|
2905 |
+
"eval_steps_per_second": 7.62,
|
2906 |
+
"step": 4200
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 6.59,
|
2910 |
+
"learning_rate": 3.528951486697966e-05,
|
2911 |
+
"loss": 0.0046,
|
2912 |
+
"step": 4210
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 6.6,
|
2916 |
+
"learning_rate": 3.489827856025039e-05,
|
2917 |
+
"loss": 0.0055,
|
2918 |
+
"step": 4220
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 6.62,
|
2922 |
+
"learning_rate": 3.450704225352113e-05,
|
2923 |
+
"loss": 0.0026,
|
2924 |
+
"step": 4230
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 6.64,
|
2928 |
+
"learning_rate": 3.411580594679186e-05,
|
2929 |
+
"loss": 0.0028,
|
2930 |
+
"step": 4240
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 6.65,
|
2934 |
+
"learning_rate": 3.3724569640062596e-05,
|
2935 |
+
"loss": 0.005,
|
2936 |
+
"step": 4250
|
2937 |
+
},
|
2938 |
+
{
|
2939 |
+
"epoch": 6.67,
|
2940 |
+
"learning_rate": 3.3333333333333335e-05,
|
2941 |
+
"loss": 0.0028,
|
2942 |
+
"step": 4260
|
2943 |
+
},
|
2944 |
+
{
|
2945 |
+
"epoch": 6.68,
|
2946 |
+
"learning_rate": 3.294209702660407e-05,
|
2947 |
+
"loss": 0.0023,
|
2948 |
+
"step": 4270
|
2949 |
+
},
|
2950 |
+
{
|
2951 |
+
"epoch": 6.7,
|
2952 |
+
"learning_rate": 3.255086071987481e-05,
|
2953 |
+
"loss": 0.0026,
|
2954 |
+
"step": 4280
|
2955 |
+
},
|
2956 |
+
{
|
2957 |
+
"epoch": 6.71,
|
2958 |
+
"learning_rate": 3.215962441314554e-05,
|
2959 |
+
"loss": 0.0025,
|
2960 |
+
"step": 4290
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 6.73,
|
2964 |
+
"learning_rate": 3.176838810641627e-05,
|
2965 |
+
"loss": 0.0031,
|
2966 |
+
"step": 4300
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 6.73,
|
2970 |
+
"eval_accuracy": 0.9113338473400154,
|
2971 |
+
"eval_loss": 0.4005250632762909,
|
2972 |
+
"eval_runtime": 43.1239,
|
2973 |
+
"eval_samples_per_second": 60.152,
|
2974 |
+
"eval_steps_per_second": 7.536,
|
2975 |
+
"step": 4300
|
2976 |
+
},
|
2977 |
+
{
|
2978 |
+
"epoch": 6.74,
|
2979 |
+
"learning_rate": 3.137715179968701e-05,
|
2980 |
+
"loss": 0.0027,
|
2981 |
+
"step": 4310
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 6.76,
|
2985 |
+
"learning_rate": 3.0985915492957744e-05,
|
2986 |
+
"loss": 0.0024,
|
2987 |
+
"step": 4320
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 6.78,
|
2991 |
+
"learning_rate": 3.0594679186228484e-05,
|
2992 |
+
"loss": 0.0031,
|
2993 |
+
"step": 4330
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 6.79,
|
2997 |
+
"learning_rate": 3.0203442879499216e-05,
|
2998 |
+
"loss": 0.0109,
|
2999 |
+
"step": 4340
|
3000 |
+
},
|
3001 |
+
{
|
3002 |
+
"epoch": 6.81,
|
3003 |
+
"learning_rate": 2.9812206572769952e-05,
|
3004 |
+
"loss": 0.0032,
|
3005 |
+
"step": 4350
|
3006 |
+
},
|
3007 |
+
{
|
3008 |
+
"epoch": 6.82,
|
3009 |
+
"learning_rate": 2.9420970266040692e-05,
|
3010 |
+
"loss": 0.0027,
|
3011 |
+
"step": 4360
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 6.84,
|
3015 |
+
"learning_rate": 2.9029733959311428e-05,
|
3016 |
+
"loss": 0.0033,
|
3017 |
+
"step": 4370
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 6.85,
|
3021 |
+
"learning_rate": 2.8638497652582164e-05,
|
3022 |
+
"loss": 0.0044,
|
3023 |
+
"step": 4380
|
3024 |
+
},
|
3025 |
+
{
|
3026 |
+
"epoch": 6.87,
|
3027 |
+
"learning_rate": 2.82472613458529e-05,
|
3028 |
+
"loss": 0.0029,
|
3029 |
+
"step": 4390
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 6.89,
|
3033 |
+
"learning_rate": 2.7856025039123636e-05,
|
3034 |
+
"loss": 0.0021,
|
3035 |
+
"step": 4400
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 6.89,
|
3039 |
+
"eval_accuracy": 0.912875867386276,
|
3040 |
+
"eval_loss": 0.40083467960357666,
|
3041 |
+
"eval_runtime": 43.0173,
|
3042 |
+
"eval_samples_per_second": 60.301,
|
3043 |
+
"eval_steps_per_second": 7.555,
|
3044 |
+
"step": 4400
|
3045 |
+
},
|
3046 |
+
{
|
3047 |
+
"epoch": 6.9,
|
3048 |
+
"learning_rate": 2.746478873239437e-05,
|
3049 |
+
"loss": 0.0021,
|
3050 |
+
"step": 4410
|
3051 |
+
},
|
3052 |
+
{
|
3053 |
+
"epoch": 6.92,
|
3054 |
+
"learning_rate": 2.7073552425665105e-05,
|
3055 |
+
"loss": 0.0031,
|
3056 |
+
"step": 4420
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 6.93,
|
3060 |
+
"learning_rate": 2.668231611893584e-05,
|
3061 |
+
"loss": 0.0023,
|
3062 |
+
"step": 4430
|
3063 |
+
},
|
3064 |
+
{
|
3065 |
+
"epoch": 6.95,
|
3066 |
+
"learning_rate": 2.6291079812206577e-05,
|
3067 |
+
"loss": 0.0032,
|
3068 |
+
"step": 4440
|
3069 |
+
},
|
3070 |
+
{
|
3071 |
+
"epoch": 6.96,
|
3072 |
+
"learning_rate": 2.5899843505477313e-05,
|
3073 |
+
"loss": 0.0339,
|
3074 |
+
"step": 4450
|
3075 |
+
},
|
3076 |
+
{
|
3077 |
+
"epoch": 6.98,
|
3078 |
+
"learning_rate": 2.5508607198748045e-05,
|
3079 |
+
"loss": 0.0025,
|
3080 |
+
"step": 4460
|
3081 |
+
},
|
3082 |
+
{
|
3083 |
+
"epoch": 7.0,
|
3084 |
+
"learning_rate": 2.511737089201878e-05,
|
3085 |
+
"loss": 0.0057,
|
3086 |
+
"step": 4470
|
3087 |
+
},
|
3088 |
+
{
|
3089 |
+
"epoch": 7.01,
|
3090 |
+
"learning_rate": 2.4726134585289514e-05,
|
3091 |
+
"loss": 0.0029,
|
3092 |
+
"step": 4480
|
3093 |
+
},
|
3094 |
+
{
|
3095 |
+
"epoch": 7.03,
|
3096 |
+
"learning_rate": 2.433489827856025e-05,
|
3097 |
+
"loss": 0.0019,
|
3098 |
+
"step": 4490
|
3099 |
+
},
|
3100 |
+
{
|
3101 |
+
"epoch": 7.04,
|
3102 |
+
"learning_rate": 2.3943661971830986e-05,
|
3103 |
+
"loss": 0.0021,
|
3104 |
+
"step": 4500
|
3105 |
+
},
|
3106 |
+
{
|
3107 |
+
"epoch": 7.04,
|
3108 |
+
"eval_accuracy": 0.9113338473400154,
|
3109 |
+
"eval_loss": 0.3966985046863556,
|
3110 |
+
"eval_runtime": 43.1598,
|
3111 |
+
"eval_samples_per_second": 60.102,
|
3112 |
+
"eval_steps_per_second": 7.53,
|
3113 |
+
"step": 4500
|
3114 |
+
},
|
3115 |
+
{
|
3116 |
+
"epoch": 7.06,
|
3117 |
+
"learning_rate": 2.3552425665101726e-05,
|
3118 |
+
"loss": 0.0034,
|
3119 |
+
"step": 4510
|
3120 |
+
},
|
3121 |
+
{
|
3122 |
+
"epoch": 7.07,
|
3123 |
+
"learning_rate": 2.3161189358372458e-05,
|
3124 |
+
"loss": 0.0023,
|
3125 |
+
"step": 4520
|
3126 |
+
},
|
3127 |
+
{
|
3128 |
+
"epoch": 7.09,
|
3129 |
+
"learning_rate": 2.2769953051643194e-05,
|
3130 |
+
"loss": 0.0025,
|
3131 |
+
"step": 4530
|
3132 |
+
},
|
3133 |
+
{
|
3134 |
+
"epoch": 7.1,
|
3135 |
+
"learning_rate": 2.237871674491393e-05,
|
3136 |
+
"loss": 0.0021,
|
3137 |
+
"step": 4540
|
3138 |
+
},
|
3139 |
+
{
|
3140 |
+
"epoch": 7.12,
|
3141 |
+
"learning_rate": 2.1987480438184666e-05,
|
3142 |
+
"loss": 0.0025,
|
3143 |
+
"step": 4550
|
3144 |
+
},
|
3145 |
+
{
|
3146 |
+
"epoch": 7.14,
|
3147 |
+
"learning_rate": 2.1596244131455402e-05,
|
3148 |
+
"loss": 0.0025,
|
3149 |
+
"step": 4560
|
3150 |
+
},
|
3151 |
+
{
|
3152 |
+
"epoch": 7.15,
|
3153 |
+
"learning_rate": 2.1205007824726135e-05,
|
3154 |
+
"loss": 0.0025,
|
3155 |
+
"step": 4570
|
3156 |
+
},
|
3157 |
+
{
|
3158 |
+
"epoch": 7.17,
|
3159 |
+
"learning_rate": 2.081377151799687e-05,
|
3160 |
+
"loss": 0.0023,
|
3161 |
+
"step": 4580
|
3162 |
+
},
|
3163 |
+
{
|
3164 |
+
"epoch": 7.18,
|
3165 |
+
"learning_rate": 2.0422535211267607e-05,
|
3166 |
+
"loss": 0.0023,
|
3167 |
+
"step": 4590
|
3168 |
+
},
|
3169 |
+
{
|
3170 |
+
"epoch": 7.2,
|
3171 |
+
"learning_rate": 2.0031298904538343e-05,
|
3172 |
+
"loss": 0.0043,
|
3173 |
+
"step": 4600
|
3174 |
+
},
|
3175 |
+
{
|
3176 |
+
"epoch": 7.2,
|
3177 |
+
"eval_accuracy": 0.9121048573631457,
|
3178 |
+
"eval_loss": 0.3959566652774811,
|
3179 |
+
"eval_runtime": 43.6577,
|
3180 |
+
"eval_samples_per_second": 59.417,
|
3181 |
+
"eval_steps_per_second": 7.444,
|
3182 |
+
"step": 4600
|
3183 |
+
},
|
3184 |
+
{
|
3185 |
+
"epoch": 7.21,
|
3186 |
+
"learning_rate": 1.964006259780908e-05,
|
3187 |
+
"loss": 0.0021,
|
3188 |
+
"step": 4610
|
3189 |
+
},
|
3190 |
+
{
|
3191 |
+
"epoch": 7.23,
|
3192 |
+
"learning_rate": 1.9248826291079812e-05,
|
3193 |
+
"loss": 0.0024,
|
3194 |
+
"step": 4620
|
3195 |
+
},
|
3196 |
+
{
|
3197 |
+
"epoch": 7.25,
|
3198 |
+
"learning_rate": 1.8857589984350548e-05,
|
3199 |
+
"loss": 0.0023,
|
3200 |
+
"step": 4630
|
3201 |
+
},
|
3202 |
+
{
|
3203 |
+
"epoch": 7.26,
|
3204 |
+
"learning_rate": 1.8466353677621284e-05,
|
3205 |
+
"loss": 0.0019,
|
3206 |
+
"step": 4640
|
3207 |
+
},
|
3208 |
+
{
|
3209 |
+
"epoch": 7.28,
|
3210 |
+
"learning_rate": 1.807511737089202e-05,
|
3211 |
+
"loss": 0.002,
|
3212 |
+
"step": 4650
|
3213 |
+
},
|
3214 |
+
{
|
3215 |
+
"epoch": 7.29,
|
3216 |
+
"learning_rate": 1.7683881064162756e-05,
|
3217 |
+
"loss": 0.0023,
|
3218 |
+
"step": 4660
|
3219 |
+
},
|
3220 |
+
{
|
3221 |
+
"epoch": 7.31,
|
3222 |
+
"learning_rate": 1.7292644757433492e-05,
|
3223 |
+
"loss": 0.0022,
|
3224 |
+
"step": 4670
|
3225 |
+
},
|
3226 |
+
{
|
3227 |
+
"epoch": 7.32,
|
3228 |
+
"learning_rate": 1.6901408450704224e-05,
|
3229 |
+
"loss": 0.0021,
|
3230 |
+
"step": 4680
|
3231 |
+
},
|
3232 |
+
{
|
3233 |
+
"epoch": 7.34,
|
3234 |
+
"learning_rate": 1.651017214397496e-05,
|
3235 |
+
"loss": 0.0021,
|
3236 |
+
"step": 4690
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 7.36,
|
3240 |
+
"learning_rate": 1.6118935837245697e-05,
|
3241 |
+
"loss": 0.0022,
|
3242 |
+
"step": 4700
|
3243 |
+
},
|
3244 |
+
{
|
3245 |
+
"epoch": 7.36,
|
3246 |
+
"eval_accuracy": 0.9124903623747108,
|
3247 |
+
"eval_loss": 0.3962063193321228,
|
3248 |
+
"eval_runtime": 43.4452,
|
3249 |
+
"eval_samples_per_second": 59.707,
|
3250 |
+
"eval_steps_per_second": 7.481,
|
3251 |
+
"step": 4700
|
3252 |
+
},
|
3253 |
+
{
|
3254 |
+
"epoch": 7.37,
|
3255 |
+
"learning_rate": 1.5727699530516433e-05,
|
3256 |
+
"loss": 0.0023,
|
3257 |
+
"step": 4710
|
3258 |
+
},
|
3259 |
+
{
|
3260 |
+
"epoch": 7.39,
|
3261 |
+
"learning_rate": 1.533646322378717e-05,
|
3262 |
+
"loss": 0.0028,
|
3263 |
+
"step": 4720
|
3264 |
+
},
|
3265 |
+
{
|
3266 |
+
"epoch": 7.4,
|
3267 |
+
"learning_rate": 1.4945226917057903e-05,
|
3268 |
+
"loss": 0.0022,
|
3269 |
+
"step": 4730
|
3270 |
+
},
|
3271 |
+
{
|
3272 |
+
"epoch": 7.42,
|
3273 |
+
"learning_rate": 1.4553990610328639e-05,
|
3274 |
+
"loss": 0.0023,
|
3275 |
+
"step": 4740
|
3276 |
+
},
|
3277 |
+
{
|
3278 |
+
"epoch": 7.43,
|
3279 |
+
"learning_rate": 1.4162754303599373e-05,
|
3280 |
+
"loss": 0.0025,
|
3281 |
+
"step": 4750
|
3282 |
+
},
|
3283 |
+
{
|
3284 |
+
"epoch": 7.45,
|
3285 |
+
"learning_rate": 1.377151799687011e-05,
|
3286 |
+
"loss": 0.013,
|
3287 |
+
"step": 4760
|
3288 |
+
},
|
3289 |
+
{
|
3290 |
+
"epoch": 7.46,
|
3291 |
+
"learning_rate": 1.3380281690140845e-05,
|
3292 |
+
"loss": 0.0021,
|
3293 |
+
"step": 4770
|
3294 |
+
},
|
3295 |
+
{
|
3296 |
+
"epoch": 7.48,
|
3297 |
+
"learning_rate": 1.298904538341158e-05,
|
3298 |
+
"loss": 0.0023,
|
3299 |
+
"step": 4780
|
3300 |
+
},
|
3301 |
+
{
|
3302 |
+
"epoch": 7.5,
|
3303 |
+
"learning_rate": 1.2597809076682316e-05,
|
3304 |
+
"loss": 0.0027,
|
3305 |
+
"step": 4790
|
3306 |
+
},
|
3307 |
+
{
|
3308 |
+
"epoch": 7.51,
|
3309 |
+
"learning_rate": 1.2206572769953052e-05,
|
3310 |
+
"loss": 0.0021,
|
3311 |
+
"step": 4800
|
3312 |
+
},
|
3313 |
+
{
|
3314 |
+
"epoch": 7.51,
|
3315 |
+
"eval_accuracy": 0.9121048573631457,
|
3316 |
+
"eval_loss": 0.39916661381721497,
|
3317 |
+
"eval_runtime": 43.2453,
|
3318 |
+
"eval_samples_per_second": 59.983,
|
3319 |
+
"eval_steps_per_second": 7.515,
|
3320 |
+
"step": 4800
|
3321 |
+
},
|
3322 |
+
{
|
3323 |
+
"epoch": 7.53,
|
3324 |
+
"learning_rate": 1.1815336463223788e-05,
|
3325 |
+
"loss": 0.0021,
|
3326 |
+
"step": 4810
|
3327 |
+
},
|
3328 |
+
{
|
3329 |
+
"epoch": 7.54,
|
3330 |
+
"learning_rate": 1.1424100156494522e-05,
|
3331 |
+
"loss": 0.002,
|
3332 |
+
"step": 4820
|
3333 |
+
},
|
3334 |
+
{
|
3335 |
+
"epoch": 7.56,
|
3336 |
+
"learning_rate": 1.1032863849765258e-05,
|
3337 |
+
"loss": 0.0019,
|
3338 |
+
"step": 4830
|
3339 |
+
},
|
3340 |
+
{
|
3341 |
+
"epoch": 7.57,
|
3342 |
+
"learning_rate": 1.0641627543035994e-05,
|
3343 |
+
"loss": 0.0025,
|
3344 |
+
"step": 4840
|
3345 |
+
},
|
3346 |
+
{
|
3347 |
+
"epoch": 7.59,
|
3348 |
+
"learning_rate": 1.0250391236306729e-05,
|
3349 |
+
"loss": 0.0023,
|
3350 |
+
"step": 4850
|
3351 |
+
},
|
3352 |
+
{
|
3353 |
+
"epoch": 7.61,
|
3354 |
+
"learning_rate": 9.859154929577465e-06,
|
3355 |
+
"loss": 0.0022,
|
3356 |
+
"step": 4860
|
3357 |
+
},
|
3358 |
+
{
|
3359 |
+
"epoch": 7.62,
|
3360 |
+
"learning_rate": 9.4679186228482e-06,
|
3361 |
+
"loss": 0.0108,
|
3362 |
+
"step": 4870
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 7.64,
|
3366 |
+
"learning_rate": 9.076682316118937e-06,
|
3367 |
+
"loss": 0.0021,
|
3368 |
+
"step": 4880
|
3369 |
+
},
|
3370 |
+
{
|
3371 |
+
"epoch": 7.65,
|
3372 |
+
"learning_rate": 8.685446009389673e-06,
|
3373 |
+
"loss": 0.0023,
|
3374 |
+
"step": 4890
|
3375 |
+
},
|
3376 |
+
{
|
3377 |
+
"epoch": 7.67,
|
3378 |
+
"learning_rate": 8.294209702660407e-06,
|
3379 |
+
"loss": 0.002,
|
3380 |
+
"step": 4900
|
3381 |
+
},
|
3382 |
+
{
|
3383 |
+
"epoch": 7.67,
|
3384 |
+
"eval_accuracy": 0.912875867386276,
|
3385 |
+
"eval_loss": 0.39511793851852417,
|
3386 |
+
"eval_runtime": 43.4397,
|
3387 |
+
"eval_samples_per_second": 59.715,
|
3388 |
+
"eval_steps_per_second": 7.482,
|
3389 |
+
"step": 4900
|
3390 |
+
},
|
3391 |
+
{
|
3392 |
+
"epoch": 7.68,
|
3393 |
+
"learning_rate": 7.902973395931143e-06,
|
3394 |
+
"loss": 0.0024,
|
3395 |
+
"step": 4910
|
3396 |
+
},
|
3397 |
+
{
|
3398 |
+
"epoch": 7.7,
|
3399 |
+
"learning_rate": 7.511737089201878e-06,
|
3400 |
+
"loss": 0.0021,
|
3401 |
+
"step": 4920
|
3402 |
+
},
|
3403 |
+
{
|
3404 |
+
"epoch": 7.72,
|
3405 |
+
"learning_rate": 7.120500782472614e-06,
|
3406 |
+
"loss": 0.0019,
|
3407 |
+
"step": 4930
|
3408 |
+
},
|
3409 |
+
{
|
3410 |
+
"epoch": 7.73,
|
3411 |
+
"learning_rate": 6.7292644757433494e-06,
|
3412 |
+
"loss": 0.0022,
|
3413 |
+
"step": 4940
|
3414 |
+
},
|
3415 |
+
{
|
3416 |
+
"epoch": 7.75,
|
3417 |
+
"learning_rate": 6.338028169014085e-06,
|
3418 |
+
"loss": 0.0023,
|
3419 |
+
"step": 4950
|
3420 |
+
},
|
3421 |
+
{
|
3422 |
+
"epoch": 7.76,
|
3423 |
+
"learning_rate": 5.94679186228482e-06,
|
3424 |
+
"loss": 0.002,
|
3425 |
+
"step": 4960
|
3426 |
+
},
|
3427 |
+
{
|
3428 |
+
"epoch": 7.78,
|
3429 |
+
"learning_rate": 5.555555555555556e-06,
|
3430 |
+
"loss": 0.0017,
|
3431 |
+
"step": 4970
|
3432 |
+
},
|
3433 |
+
{
|
3434 |
+
"epoch": 7.79,
|
3435 |
+
"learning_rate": 5.164319248826292e-06,
|
3436 |
+
"loss": 0.0024,
|
3437 |
+
"step": 4980
|
3438 |
+
},
|
3439 |
+
{
|
3440 |
+
"epoch": 7.81,
|
3441 |
+
"learning_rate": 4.773082942097027e-06,
|
3442 |
+
"loss": 0.0024,
|
3443 |
+
"step": 4990
|
3444 |
+
},
|
3445 |
+
{
|
3446 |
+
"epoch": 7.82,
|
3447 |
+
"learning_rate": 4.381846635367762e-06,
|
3448 |
+
"loss": 0.0023,
|
3449 |
+
"step": 5000
|
3450 |
+
},
|
3451 |
+
{
|
3452 |
+
"epoch": 7.82,
|
3453 |
+
"eval_accuracy": 0.9124903623747108,
|
3454 |
+
"eval_loss": 0.3951539695262909,
|
3455 |
+
"eval_runtime": 42.332,
|
3456 |
+
"eval_samples_per_second": 61.278,
|
3457 |
+
"eval_steps_per_second": 7.677,
|
3458 |
+
"step": 5000
|
3459 |
+
},
|
3460 |
+
{
|
3461 |
+
"epoch": 7.84,
|
3462 |
+
"learning_rate": 3.990610328638498e-06,
|
3463 |
+
"loss": 0.0021,
|
3464 |
+
"step": 5010
|
3465 |
+
},
|
3466 |
+
{
|
3467 |
+
"epoch": 7.86,
|
3468 |
+
"learning_rate": 3.5993740219092334e-06,
|
3469 |
+
"loss": 0.0025,
|
3470 |
+
"step": 5020
|
3471 |
+
},
|
3472 |
+
{
|
3473 |
+
"epoch": 7.87,
|
3474 |
+
"learning_rate": 3.2081377151799686e-06,
|
3475 |
+
"loss": 0.0029,
|
3476 |
+
"step": 5030
|
3477 |
+
},
|
3478 |
+
{
|
3479 |
+
"epoch": 7.89,
|
3480 |
+
"learning_rate": 2.8169014084507042e-06,
|
3481 |
+
"loss": 0.0024,
|
3482 |
+
"step": 5040
|
3483 |
+
},
|
3484 |
+
{
|
3485 |
+
"epoch": 7.9,
|
3486 |
+
"learning_rate": 2.42566510172144e-06,
|
3487 |
+
"loss": 0.0019,
|
3488 |
+
"step": 5050
|
3489 |
+
},
|
3490 |
+
{
|
3491 |
+
"epoch": 7.92,
|
3492 |
+
"learning_rate": 2.0344287949921754e-06,
|
3493 |
+
"loss": 0.0022,
|
3494 |
+
"step": 5060
|
3495 |
+
},
|
3496 |
+
{
|
3497 |
+
"epoch": 7.93,
|
3498 |
+
"learning_rate": 1.643192488262911e-06,
|
3499 |
+
"loss": 0.0025,
|
3500 |
+
"step": 5070
|
3501 |
+
},
|
3502 |
+
{
|
3503 |
+
"epoch": 7.95,
|
3504 |
+
"learning_rate": 1.2519561815336464e-06,
|
3505 |
+
"loss": 0.0018,
|
3506 |
+
"step": 5080
|
3507 |
+
},
|
3508 |
+
{
|
3509 |
+
"epoch": 7.97,
|
3510 |
+
"learning_rate": 8.607198748043818e-07,
|
3511 |
+
"loss": 0.0017,
|
3512 |
+
"step": 5090
|
3513 |
+
},
|
3514 |
+
{
|
3515 |
+
"epoch": 7.98,
|
3516 |
+
"learning_rate": 4.694835680751174e-07,
|
3517 |
+
"loss": 0.0021,
|
3518 |
+
"step": 5100
|
3519 |
+
},
|
3520 |
+
{
|
3521 |
+
"epoch": 7.98,
|
3522 |
+
"eval_accuracy": 0.912875867386276,
|
3523 |
+
"eval_loss": 0.39521878957748413,
|
3524 |
+
"eval_runtime": 43.0806,
|
3525 |
+
"eval_samples_per_second": 60.213,
|
3526 |
+
"eval_steps_per_second": 7.544,
|
3527 |
+
"step": 5100
|
3528 |
+
},
|
3529 |
+
{
|
3530 |
+
"epoch": 8.0,
|
3531 |
+
"learning_rate": 7.82472613458529e-08,
|
3532 |
+
"loss": 0.0026,
|
3533 |
+
"step": 5110
|
3534 |
+
},
|
3535 |
+
{
|
3536 |
+
"epoch": 8.0,
|
3537 |
+
"step": 5112,
|
3538 |
+
"total_flos": 6.337884979995771e+18,
|
3539 |
+
"train_loss": 0.4943873079753071,
|
3540 |
+
"train_runtime": 5596.588,
|
3541 |
+
"train_samples_per_second": 14.595,
|
3542 |
+
"train_steps_per_second": 0.913
|
3543 |
+
}
|
3544 |
+
],
|
3545 |
+
"max_steps": 5112,
|
3546 |
+
"num_train_epochs": 8,
|
3547 |
+
"total_flos": 6.337884979995771e+18,
|
3548 |
+
"trial_name": null,
|
3549 |
+
"trial_params": null
|
3550 |
+
}
|